Publicly accessible
URL: http://nzetc.victoria.ac.nz/collections.html
copyright 2006, by Victoria University of Wellington
All unambiguous end-of-line hyphens have been removed and the trailing part of a word has been joined to the preceding line, except in the case of those words that break over a page. Every effort has been made to preserve the Māori macron using unicode.
Some keywords in the header are a local Electronic Text Center scheme to aid in establishing analytical groupings.
Campbell, Ella O.
Fleming, C. A.
Johnston, H. W.
McQueen, D. R.
Ottaway, J. R.
ALGAE—
ARCTOCEPHALUS FORSTERI—
CULTURING ALGAE—See ALGAE
DISTRIBUTION—
ECOLOGY—See NOTHOFAGUS
LETTER—
LIVERWORTS—See RICCIACEAE
MARINE BIOLOGY—
NOTHOFAGUS—
PREDATORS—
REGENERATION—See NOTHOFAGUS
REVIEWS—
RICCIACEAE—
SEA ANEMONES—See PREDATORS
SEAL, FUR—See ARCTOCEPHALUS FORSTERI.
The New Zealand fur seal, Arctocephalus forsteri (Lesson, 1828), lives and breeds on the rocky shores of southern New Zealand and its subantarctic islands. Although brought close to extinction by commercial sealers during the nineteenth century, the species has recovered well under protection and is increasing steadily in numbers and expanding its range (Csordas and Ingham, 1965; Falla, 1965; Stonehouse, 1965; Stirling, 1968). The fur seal was first given legal protection in 1875, and from then until 1916 seasons were of limited length and permits were required for seal killing. Since 1916 no fur seals have been legally killed (except for research purposes) apart from on Campbell Island in 1924 and 1926, and in parts of southern New Zealand in 1946 (Sorensen, 1969b).
By comparison with most other species of otariid seal the biology of A. forsteri is poorly known, but there have been rapid advances in knowledge in recent years. In particular, since 1969 there have been intensive studies of distribution, abundance, population structure, breeding biology and behaviour, and some results have already been published (Stirling, 1970; Crawley and Brown, 1971; Miller, 1971, 1974; Crawley, 1972, in press; McNab and Crawley, in press; Wilson 1974a, b). The new information renders existing general accounts of the biology of A. forsteri unsatisfactory in some respects (Scheffer, 1958; King, 1964; Sorensen, 1969a, b; Gaskin, 1972), so the aim here is to present a more up to date account of the natural history and behaviour of the New Zealand fur seal. In this paper, the emphasis is on presenting the most recent information on the topics covered, rather than on reviewing the literature, but we hope that no important material has been omitted. The account relies a great deal on our own work on distribution, breeding biology and social behaviour, the full results of which will be published separately. The observational
Taxonomic confusion within the genus Arctocephalus has been clarified by King (1968, 1969), Shaughnessy (1970), Repenning et al. (1971) and Stirling and Warneke (1971). The current view is that there are eight species of Arctocephalus, defined chiefly by characteristics of the skull. The decision by Repenning et al. to synonymise the New Zealand fur seal, A. forsteri, and the South Australian fur seal, formerly A. doriferus (Wood Jones), is of particular relevance to this paper. Work on the South Australian fur seal by Stirling (1971a, b) and Stirling and Warneke (1971) adds considerably to our knowledge of the social behaviour of A. forsteri, but because of the unknown effects of geographic separation and climatic differences between New Zealand and South Australia, only New Zealand populations are discussed here unless otherwise stated.
The work of Repenning et al. (1971) shows that there are three sizes of fur seals: a large species, A. pusillus; a small species, A. galapagoensis; and a group of six medium-sized species, including A. forsteri. A. forsteri has been described by Webb (1871), Clark (1875), Thomson (1921) and Sivertson (1954), and fuller details of appearance and morphology can be found in these papers. The coat colour of adults merges from a dark grey-brown dorsally to a lighter grey-brown ventrally. The thick underfur is a rich chestnut colour; the guard hairs are coarse and dark grey, often with white tips which impart a silvery sheen to the dry fur. Bulls have thick manes composed of long, coarse guard hairs. Pups are black at birth but moult to a silvery-grey after about five months. The fur appears darker when wet. Adults and pups are illustrated in Figs. 1-3.
Adult males (bulls) reach 2 m and 200 kg and adult females (cows) about 1.5 m and 90 kg. Bulls are much more massive than cows, particularly around the neck and shoulders, and during the breeding season the average male: female weight ratio is 3.5: 1.
At birth, pups are about 55 cm long and weigh 3.5 kg. At about six weeks old, male pups (mean = 5.1 kg) are significantly heavier (P < 0.001) than females (mean = 4.5 kg), but are not significantly longer (66.6 cm and 64.5 cm respectively; Crawley and Brown, 1971). From birth to 240 days, both sexes gain an average of 24 g and 0.86 cm per day (Crawley, in press).
The New Zealand fur seal, like the South African fur seal, A. p. pusillus (Rand, 1967), is a coastal species with a rather limited range. Distant migrations are not performed and it is probable that the bulk of the population remains near land.
The present distribution of New Zealand A. forsteri, as determined from published information, questionnaires and personal observations, is shown in Figs. 4 and 5. Fur seals penetrate as far north as the Three Kings Islands (34°10′ S., 172°08′ E.; Singleton, 1972) and as far south as Macquarie Island (54°30′ S., 157°E.; Csordas and Ingham, 1965). Within these latitudes, fur seal distribution is discontinuous and seasonally variable. Throughout the year, seals are most plentiful on the rocky coasts of the South Island, Stewart Island, the Chatham Islands and the New Zealand subantarctic islands, but in winter large (> 500) colonies form in the south of the North Island, and small groups occupy suitable coastal terrain as far north as Three Kings Islands.
Seasonal variation in fur seal distribution is partly a consequence of the breeding regime of the species. Except for the isolated breeding colonies (rookeries) at Cape Foulwind and to the north of the Heaphy
The breeding season is from late October to early February for the territorial bulls, but cows and pups remain on the rookeries until August-September. Only in winter and early spring are seals ashore in significant numbers on the hauling grounds north of the breeding range; from May onwards adult and subadult males come ashore in increasing numbers to reach peak density in July-August. These essentially bachelor bull colonies diminish rapidly in size in September-October. On most winter haul-out sites there are always a few pre-breeder or post-reproductive males ashore, but in the far north the hauling grounds may be deserted during the summer. During the breeding season the fur seal distribution is more restricted than at other times of the year, but obviously not all the seals are involved in breeding. Non-breeders congregate in large numbers on rocky beaches adjacent to the rookeries or on nearby islets. On many islands, including the Chatham Islands, Big Solander Island, Campbell Island (Bailey and Sorensen, 1962), Auckland Islands (Wilson, 1974a) and the Snares Islands (Crawley, 1972), both non-breeding colonies and rookeries occur. The rookeries are usually on the exposed west coasts of the islands, while the hauling grounds are on the more sheltered east coasts. Campbell Island seals show a reversal of this tendency, due probably to an absence of suitable habitats. In addition to the discrete rookeries and main hauling grounds, there are concentrations of non-breeders, mainly large subadult males, in the vicinity of rookeries.
Characteristically, the fur seal is an inhabitant of exposed rocky coastline, but there are records of groups of seals in sheltered waterways. There is a small, but well-established, bachelor bull colony in Paterson Inlet, Stewart Island, and seals frequently haul out on islands in Dusky Sound and in Milford Sound. Seals have also been recorded from most of the fiords and sounds, from rivers in Otago, Canterbury, Marlborough and Nelson, and from Otago and Hokianga Harbours.
Fur seal distribution changes seasonally, mainly because of the movements of males. Females appear to divide their time between the rookery (intermittent presence between November and August) and the sea, whereas males and subadults move between rookeries and hauling grounds and come ashore at suitable locations en route. Knowledge of the seasonal movements is limited to what can be deduced from counts at various localities at different times of the year.
On the east coast of New Zealand repetitive counts are available from Bunkers Islets (46° 50′ S.), Tia Island (47° 05′ S.), Cape Saunders (45° 50′ S.), Motunau Island (43° 05′ S.), Kaikoura (42° 20′ S.) and Turakirae Head (41° 25′ S.). Around Stewart Island, maximum numbers of seals are ashore at breeding colonies in autumn. Fur seal numbers at Cape Saunders reach a peak in January, decline until mid-winter, increase again in August-September, and finally decline to a minimum in November. At Motunau Island (Cox et al., 1970) seals have been seen ashore in January, April, May-September and October, being most numerous in September. At Kaikoura, maximum numbers occur in May and June and numbers remain high until October, after which there is a rapid decline to a summer minimum (November-January), followed by a steady increase again in February and March (Stonehouse, 1965; Stirling, 1970; Miller, 1971). At Turakirae Head, monthly counts made by Whitaker (unpublished) from May, 1967, to August, 1969, showed maximum numbers ashore in June-July; very few were seen in October-January.
The above information suggests that there is a movement of seals northwards from Stewart Island beginning in January, with many reaching Cape Saunders in January-February, Kaikoura in May-June, and the Wellington area in July. Beginning in August, a southward movement occurs and by September many are again in the Otago area. In November, numbers ashore are at a minimum everywhere.
On the west coast the situation is different. Seasonal counts are available only for Gillespie's Point (43° 25′ S.), where maximum numbers are ashore in August-September, about two months later than at the same latitude on the east coast. This time-lag is unexplained at present.
Some seals move south from the breeding grounds in late summer. At Macquarie Island (54° 30′ S.) there is an influx of non-breeders in February-March, but most leave again before mid-winter (Csordas and Ingham, 1965). Movements of seals between the other subantarctic islands are unknown.
Fur seals of the genus Arctocephalus are nearly always found on exposed rocky coastline. Falla (in Sorensen. 1969b) recognised three major types of habitat used by New Zealand fur seals: tumbledown (talus) beaches, terraced rocky ledges and small islets. He suggested that these are used by breeding animals, non-breeding males and immatures respectively. Our work suggests that this is an oversimplification.
Breeding colonies are found mainly on exposed west or north-west coasts and characteristically have three main features:
Stirling (1971a) suggests that for South Australian A. forsteri the sea, or rock pools, must be readily available for cooling in warm weather. This is less important in the cooler climate of southern New Zealand and the subantarctic islands. However, warmer localities such as Open Bay Islands do have pools which are used by seals in warm weather.
Habitat requirements for non-breeding colonies are less stringent than for rookeries and a wide variety of rocky coast is used. The only critical factor appears to be easy access. Most preferred are shelving rocky ledges and boulder beaches (Fig. 7); rounded or small boulders and stones are generally avoided. Hauling grounds occur on suitable points, spurs, headlands, reefs and islets, and less often in bays. They are all regular stations and only wandering individuals come ashore elsewhere.
Fur seals make use of vegetated areas, particularly tussock and scrub, but these are usually adjacent to a colony. Such areas are usually retreats for cows and pups but are sometimes adjacent to a
The use of different types of terrain must be interpreted with caution, since it is often uncertain as to whether habitat use is due to availability or to terrain preference. The present population is still much smaller than that which existed prior to the sealing days (McNab, 1907), and we can only guess what use a larger population would make of what is presently less-preferred terrain. Also, the extent to which the present use of coastal areas depends on the seals' return to ancestral hauling grounds is conjectural. Burrows (1971) suggested, using evidence from soil studies and gastrolith finds, that seals once spread over the Open Bay Islands, whereas now they are confined to the coastal rocks and the fringes of the vegetation. Similar changes in habitat use probably also occurred elsewhere.
In his book on marine mammals of the New Zealand region, Gaskin (1972) noted that there was no reliable estimate of the numbers of the New Zealand population of A. forsteri because no complete census work had been carried out. Since 1970, we have been trying to rectify this omission, and
The counts were made during the period January 1970-February 1973, mainly during the summer, and in all kinds of weather at various times of days, as opportunity permitted. Wilson (1974b) has prepared a detailed record of counts at each colony visited, including correction factors for time of day, weather conditions, etc., but for the present purpose only the broad over-all picture is given. The reliability of the counts varies greatly, as some visits involved an extended period ashore, allowing a detailed study of the colony; others permitted only counts from a clifftop; while some precluded a count ashore altogether, and only a rough idea of numbers could be gained from a rocking boat offshore. Our estimates for each region are given in Table 1, together with data for Macquarie Island (Johnstone, 1972), Antipodes Island (Taylor, in Sorensen, 1969b), Campbell Island (Bailey and Sorensen, 1962) and the Bounty Islands (Falla, in Sorensen, 1969b).
We estimate that the number of fur seals within the breeding range is about 36,000, while the figure for the total New Zealand population could be as high as 40,000, if the non-breeding animals which remain on the hauling grounds north of the breeding range during the summer are included. This estimate of 40,000 is to be compared with that of about 20,000 by Falla (in Sorensen, 1969b), which was based mainly on counts in only part of the range in 1947-48, and 20,600 by Gaskin (1972), who used the best evidence available from his own counts and those of Falla, Sorensen and Taylor (in Sorensen, 1969b). Our estimate is about 100% greater than the previous estimates, but this is
No attempt is made here, in this general account of fur seal biology, to analyse in detail the results of the counts or to present data on age structure and sex ratio of the colonies. This information will be published separately by
There has been only one detailed investigation of New Zealand fur seal foods (Street, 1964), but observations by Falla (in Sorensen, 1969b), Rapson (in Sorensen, 1969b), Bailey and Sorensen (1962) and Csordas and Ingham (1965) have provided useful information. Our knowledge of the diet of A. forsteri remains sketchy, however, and further study is needed. A list of possible food items, based on published data and personal observations, is given in Table 2.
Street's (1964) study of fur seal foods was carried out to check the validity of claims by fishermen that seals were detrimentally affecting stocks of commercially important fish. His findings indicated that there was little justification for this view. He examined the stomach contents of 70 seals taken from Kaikoura Peninsula, Banks Peninsula, Cape Saunders, Nugget Point and Bench Island in Foveaux Strait, and discovered that octopus (Octopus maorum), squid (Notodarus sloanii and Sepioteuthis bilineata) and barracouta (Thyrsites
atun) made up 90.8% of the stomach contents by weight. Rapson (in Sorensen, 1969b) found food in only 25 of 91 stomachs examined; octopus or squid were present in 22 stomachs and fish (one barracouta, one greenbone (Coridodax pullus) and one perch-like fish) in only three. Observations by Falla (in Sorensen, 1969b), Bailey and Sorensen (1962), and Csordas and Ingham (1965) confirm that squid is a preferred food while fish are relatively unimportant in the diet.
Penguins are also commonly eaten by fur seals. Both Falla (in Sorensen, 1969b) and Csordas and Ingham (1965) recorded penguin feathers and other remains in stomachs, faeces or regurgitated material, while Bailey and Sorensen (1962) described how fur seals chased and devoured Rockhopper penguins (Eudyptes chrysocome) near the shores of Campbell Island. Seals resident on the subantarctic islands commonly include penguins in their diet, perhaps because penguins are more readily available than fish in those waters. Squid and octopus are still the major foods, however.
The work of Street (1964) indicates that fur seals feed principally in surface waters at night, when they take squid and surface fish. Octopus are collected from the sea floor, often by day. Seals have been reported by fishermen to bring large fish such as ling (Gerypterus blacodes) and blue cod (Parapercis colias) to the surface during the day and break them up for eating. It is probable that this very obvious activity over-emphasises the occurrence of such behaviour. Fur seals may even benefit the crayfishing industry by eating octopus, which are important predators of crayfish (Jasus sp.).
In summary, our present knowledge of the diet of the fur seal in New Zealand waters indicates that squid and octopus are the most important food items, and these are supplemented by penguins (particularly around subantarctic islands) and surface-feeding fish.
This general account of the annual cycle of A. forsteri is based on our own observations and the work of Sorensen (1969a, b), Stirling (1970) and Miller (1971).
Seals are present in the rookeries and on the hauling grounds throughout the year, although the proportions of the various sex and age classes vary seasonally. Few bulls frequent the rookeries in the March-September period, the resident population comprising small subadults, yearlings and cows. Yearling numbers decrease steadily between June and November and few remain when the first pups are born in November. From mid-October to mid-November the number of bulls ashore increases steadily, while cows arrive in large numbers in late November and throughout December. From mid-November to late December the number of bulls increases slowly. The relative absence of cows ashore in late November may be because pregnant females feed heavily at sea in the weeks before giving birth. Most cows
Pupping is from late November to mid-January with a peak in mid-December. Cows stay with their newborn pups for about ten days and during that period they mate with the nearest male, usually a territorial bull. Their first feeding trip is of three to five days and on their return they suckle the pups for two to four days. As the pups grow older the cows leave them alone for longer, and the pups congregate into unstable small groups called pods. Each pup leaves the pod to suckle its mother on her return from each feeding trip.
The decline in the numbers of bulls and cows ashore during January is a consequence of the gradual breakdown of the territorial system following the main birth and mating period. The bulls may have spent up to ten weeks ashore without food or water, and they depart for the feeding areas to regain their strength. The cows may have been ashore for two to three weeks, during which time they have given birth, mated, and suckled their young; they move away for their first feeding trip. In February there may be no bulls ashore, and possibly only half as many cows as in January. The feeding trips of the cows are longer than in December and January. As the adult population declines there is an influx of subadults and yearlings on to the main breeding rocks. Many of the subadults occupy offshore rocks during the breeding period, while some of the larger subadult males even infiltrate the edges of the rookery before being chased by the territorial bulls.
During the March-September non-breeding period the rookeries are occupied by pups, yearlings, small subadults of both sexes and females still suckling young. Many of the bulls and large numbers of subadult males move north for the winter. Cows probably spend long periods at sea feeding when they are not at the rookeries suckling their young.
The postures and calls of New Zealand A. forsteri have been described in detail by Stirling (1970) and Miller (1971), and those of South Australian A. forsteri by Stirling (1971a, b) and Stirling and Warneke (1971), and this brief account is based largely upon their descriptions. Our studies indicate that Stirling's (1971a, b) descriptions of the postures and calls of the South Australian fur seal apply also to the New Zealand fur seal, and in the following account his descriptive terms are used. A list of postures and calls, together with their possible functions and use by various classes of seal, is given in Table 3.
The agonistic behaviour of A. forsteri appears to be highly ritualised. Most of the postures and calls described are used by
In the non-breeding season, the relationship between individuals seems to be determined by a size-based dominance hierarchy, but during the breeding season the social situation is dominated by the territorial behaviour of the bulls. Generally speaking, territories are held by the bulls with the largest necks, these bulls normally being the largest over-all as well. Territorial bulls appears to be of equal status, the earlier size-based dominance being non-operative throughout the breeding season. Encounters between territory-holders are essentially meetings between bulls of similar size and status, and fighting is usually avoided by use of ritualised displays. Territorial bulls interact also with non-territorial bulls, cows, subadults and pups and employ a variety of postures, calls and movements in these encounters. In addition, all other classes interact with one another to varying degrees, using appropriate behaviours as they do so. Interactions between cows and pups are special, and some calls are unique to these classes.
Most calls and postures are common to all classes, though predominantly used by some, while others are confined to one, or a few classes. Most of the postures and calls are concerned with threat or submission and are used mainly, but not exclusively, by territorial bulls. Postures used frequently by members of both sexes and all ages include the normal sitting posture, the elimination posture for defaecation and urination (Stirling and Gentry, 1972), the alert posture (general awareness) and the submissive posture (seldom by territorial bulls). Both adult and subadult males adopt the various threat postures, but these are most often used by territorial bulls. In increasing intensity of threat the postures are: full neck display, neck-waving, oblique stare and horizontal neck stretch. The full neck display probably demonstrates the size of a bull and indicates his dominance, thus reducing conflicts. If this display is insufficient to resolve a dispute, the belligerents become positioned chest to chest and wave their heads from side to side out of phase with each other. This neck waving prolongs the threat display and might provide an attacking position. The oblique position of the head (given mutually) normally follows milder threats and usually precedes attack. The horizontal neck stretch is also a high-intensity threat display. Facing away is characteristic of subordinate males when confronted by dominant males, either after threat displays or fighting. Subordinate animals of both sexes adopt a submissive posture, with lowered head and neck, when seeking to avoid confrontation with dominants.
The open-mouth threat is given by all classes but is especially characteristic of cows. It may truly be a threat or may function as an appeasement posture also, particularly in encounters with territorial bulls. The specific posture for calling pups is restricted to cows and is characterised by an extended neck.
As with the postures described above, many of the calls uttered by A. forsteri are concerned with demonstrating status and communicating threat. The male full threat and low-intensity threat calls, the guttural challenge and the bark are given mainly by territorial bulls. All probably aid individual recognition and demonstrate territorial status, while the challenge and the bark may affirm territorial boundaries as well. The full threat call and the guttural challenge also indicate a readiness to fight, while the bark indicates sexual interest. Cows have a high-pitched guttural growl which serves as a threat to all other classes. Both sexes utter submissive squeals when threatened or defeated by dominant animals. The remaining calls are highly specific; cows whine (or occasionally moan) to attract their pups, while the pups wail in response to attract their mothers.
Subadult males are much quieter than adults; they give general moans, whimpers and growls, as well as submissive squeals when appropriate. The larger ones practise the various threat calls.
The ritualised threat and boundary displays given by territorial bulls serve to maintain and defend their territories without wastage of energy in fighting. Of course, fighting is common between bulls of similar size during territory establishment, and also between established males and challengers, but generally displays involving various postures and calls serve to maintain control of territories and subdue non-territorial bulls, cows and subadults.
Territorialism
The rookeries of the New Zealand fur seal have a more distinct organisation during the breeding season (October to February) than at other times of the year. During spring, a few bulls arrive on the breedings grounds and establish large, rather ill-defined territories. A size-based hierarchy operates, with subordinates being chased by larger dominant bulls. Later in the season, when more bulls are ashore, territories are smaller, better defined and vigorously defended, and all territorial bulls appear to be of similar status.
Territories are established prior to the arrival of the pregnant cows; some territories may never contain cows while others have large numbers. Territorial bulls attempt to influence the movements of pregnant cows and endeavour to maintain ‘harems’ within their territories, but with little success (Miller, 1974). Cows pass freely between territories although bulls may block them for up to an hour. Throughout the season bulls try to extend their territories to include areas with cows and this often leads to conflict with neighbouring bulls. Once cows have given birth to their single pup they remain with it near the birth site for about ten days; this sedentariness makes it probable that they will mate with the territory owner, rather than with
although bulls may block them for up to an hour. Throughout the season bulls try to extend their territories to include areas with cows an dthis often leads to conflict with neighbouring bulls. Once cows have give nbirth to their single pup they remain with it near the birth site for about ten days; this sedentariness makes it probable that they will mate with the territory owner, rather than with any other male, when they come into oustrous about eight days after giving birth. Cows leave on feeding trips every few days after the initial suckling period, and to re-establish contact with their pup they rely on locating it in the vicinity of the birth site. This fidelity to the birth site makes it appear that groups of cows constitute the ‘harem’ of a territory owner.
Bulls seldom voluntarily forsake their territories during the period between territory establishment (in late October or early November) and the termination of the period over which copulations take place (early January). They live in the confined space of the territories for up to 70 days, fighting, mating, sleeping and defaecating more or less in the same place. All other classes of seals except cows and pups are excluded from the territories until the end of the mating period. After the departure of many of the cows on their first feeding trips, strict territorial defence ceases and many bulls leave the rookery.
Formation of Territories
During winter and early spring the few bulls present on the breeding grounds are not territorial. However, on Open Bay Islands in August and September, individuals may occupy the same area of ground on several successive days. Miller (1971) divided bulls on the Open Bay Islands into three categories: early, intermediate and late arrivals. The first bulls appear ashore in mid-October and immediately establish large, rather ill-defined territories. Most of these early arrivals are quickly deposed, often without physical combat, by the intermediate arrivals, larger bulls who arrive later in October and in the first half of November. None of the early arrivals appears able to retain a territory long enough to gain access to cows, and it is possible that they are males attempting to breed for the first time. Generally, they successfully deter and chase large subadults but retreat from larger bulls. The intermediate arrivals establish territories which they do not leave, unless defeated in a fight, until the end of the mating period. The final category is that of the late arrivals, who must establish themselves by fighting. Some of these depose current
After the mating period, in late summer, the hostility of territorial bulls decreases. Some territory-holders extend their movements to include abandoned territories. Others depart, and their territories are taken over for a short time by small, unestablished bulls or large subadults. In late December and early January, the hostility of territory holders who remain on station is so reduced that they may be deposed, after a mild struggle, even by large subadults.
Territory Shape and Size
The choice of territory site by bulls is influenced by the topography of the rookery, the proximity of other bulls and the presence of other territories. Topography limits the size of a territory but the physical prowess of the bull fixes its boundaries. At the height of the season few territories exceed 100 m2; large territories present at the beginning of the season are whittled down as the density of bulls increases.
Territory boundaries usually follow natural irregularities of the terrain, and such natural boundaries seldom need demarcation and defence by the territory owners. Generally, boundary regions which are topographically well-demarcated change little, while poorly-defined borders between territories may shift in position. Trespasses by neighbours over poorly-defined boundaries are often tolerated by territory owners, but where a topographical irregularity defines the border, trespass is readily recognised and quickly dealt with.
Territories change in size and shape as areas are gained by fighting or opportunistic expansion, and lost by defeat or readjustment of boundaries following changes of neighbours. Where there are no obvious topographical boundaries to territories, bulls appear to respect invisible boundaries beyond lunging distance. These boundaries conform to pre-established limits, the outcome of fighting or threatening, and change if the territory owners change.
Types of Territories
The New Zealand fur seal does not appear to hold aquatic territories, although many are adjacent to the sea. There appear to be five main types of territories, not all being present at the same time.
Maintenance of Territories by Fighting
Bulls defend their territories for six to ten weeks without eating or drinking. Their stamina, strength, fighting ability, and probably also their experience and temperament, determine how successful they are in territory maintenance. Although border defence involves mainly ritualised displays and threats, serious challenges must be met by physical contact.
The successful bulls who are resident during the main breeding period often acquire their territories early in the season without need for much fighting, but heavy fighting is often necessary for them to maintain their position long enough to mate. Detailed descriptions of pre-fight behaviour and fighting are given by Miller (1971) and Stirling (1971a), but the main points are summarised here.
New arrivals which do not intend to dispute ownership of territories, but merely wish to make their way to unoccupied land. arrive discreetly, keeping their heads down, and move ashore cautiously, maintaining a flattened profile. They may lie down several times to escape attention. If challenged, they either flee, giving submissive calls, or, if cornered, give submissive whines, attempt to defend themselves and seek an escape route. Large challenger bulls behave quite differently: they haul out on rocks opposite the site to be contested and give threat calls while sitting alert. They then proceed quickly to the specific area chosen, making a vertical image readily
Fighting seems to be ritualised. Bulls meet chest-to-chest and attempt to grip their opponent on the face, neck, upper back or shoulder, and shake. They push with their chests in an attempt to unbalance one another, and alternately wave their necks from side to side. They also manoeuvre for position and make sudden lunges at the head and neck, sometimes inflicting severe wounds. The neck and shoulders are well protected by thick skin and fur, but the face, back, sides and tail region are relatively unprotected and easily torn open.
Bulls use the advantage of height whenever possible. In evenly-matched encounters, the bulls hang on to one another and push steadily, and stamina seems to decide the outcome. Fights seldom last longer than a few minutes. When one bull has had enough it ceases any aggressive behaviour and gives a submissive screech while backing away. This behaviour often, but not invariably, prevents further attacks by the winner. Victorious bulls often bite the hind flippers of retreating bulls, who receive further harsh treatment from the occupants of neighbouring territories.
Victory is accomplished by damaging an opponent, pushing him into an adjacent territory or outlasting him in strength. Infliction of damage is essential if challengers are to be deterred from further attacks; it is not enough for a territory-holder to be able to tolerate punishment. Challengers need to be exceptionally gifted to win early battles, as possession of a territory seems to give bulls confidence.
Fighting diminishes as territories are mutually recognised by neighbours, and threats and displays are normally sufficient to defend borders. However, although there is a decrease in the number of encounters once the females have arrived and settled down, there is an increase in the proportion of encounters involving fighting, due to the efforts of bulls in territories without females to oust more fortunate neighbours. However, over-all only about 30% of encounters need to be resolved by fighting.
Activity Budgets of Territorial Bulls
Fur seals, like most other animals, are involved in various activities throughout the day, and generally they divide their time between such activities in a fairly consistent way. So far, only the time budgets of territorial bull A. forsteri have been studied, mainly because this class of seal is easy to observe, being restricted in its movement, conspicuous, and uninvolved in care of the young or feeding. Stirling (1971a) studied activity budgets of Australian A. forsteri bulls and Miller (1971) did the same for New Zealand A. forsteri; their findings agree quite closely and the following account applies to A. forsteri territorial bulls generally.
Although the exact partitioning of time between activities varies between individuals, from day to day, and seasonally, there are nonetheless several general statements which may be made about male activities. Territorial bulls are least active during the middle of the day, and over-all they spend about 70% of their time lying down, either asleep or awake. When not sprawled out on the rocks they spend most of the rest of their time (up to 20%) in an upright posture, which combines general alertness with a normal sitting posture. The adoption of this posture is most prevalent during the peak reproductive period.
The only other activities which are important enough to single out as significant components of the activity regime are intra-sexual behaviour (displays and fighting) and inter-sexual behaviour (herding and copulating), which take up about 5% of the time. These are particularly subject to variation throughout the breeding period; in particular, bulls interact most with cows when the latter are in oestrous. Over-all, Miller noted that on Open Bay Islands, between December 6, 1970, and January 9, 1971, 87% of the interactions of territorial bulls were with cows; the remaining interactions were with other territorial bulls (9.6%), subadult males (1.8%) and pups (1.6%). Clearly, although pups are the dominant class numerically for much of the season, territorial bulls virtually ignore them.
There is still a great deal to learn about the activity budgets of territorial bulls, and the budgets of cows, subadults and pups are virtually unknown. The apparently continual activity of a seal rookery is obviously not largely caused by the actions of territorial bulls, which seek to conserve energy by eliminating unnecessary movements, so the other classes of seal present, particularly pups, must be the main contributors.
Copulations
Copulations may occur between mid-November and mid-January, but most are in December because of the post-partum oestrous exhibited by cows. Mating early in the season probably involves virgin cows or some who failed to rear young the previous year. Cows with pups come into oestrous about eight days after giving birth and normally, but not invariably, mate with the owner of the territory.
Bulls continually monitor the sexual state of the cows in their territory by sniffing their perineal region (Miller, 1974). Receptive cows are playful and non-aggressive towards bulls, but there is no true courtship. Bulls whimper and seek body contact on recognising a receptive cow. Mounting is dorsoventral and is often repeated. Bulls frequently bite cows during copulation, the resulting activation of the cow apparently exciting them. Cows show increased aggression after a time and bulls have to work hard to keep them in position until ejaculation has occurred. Copulations take from five to 30 minutes.
Dispersion of Breeding Cows
The dispersion of breeding cows affects that of territorial bulls. Cows arrive on the rookeries in greatest numbers in late November and early December. They prefer certain locations, usually within the seaward territories, and tolerate a certain amount of crowding rather than move on to flat, relatively featureless areas. There is some avoidance by newcomers of very crowded areas where ownership of resting ledges is disputed.
Most births take place near some sort of topographical irregularity such as a boulder, cliffside or washed-up log. Certain areas, particularly rock-filled guts, are preferred for pupping, and are used by a succession of cows. As post-parturient cows and their pups remain close to the birth spot for about ten days, cow and pup numbers increase in the preferred areas. Some competition for space probably results in the gradual use of pupping sites further inland by cows pupping later, but the complete lack of use of some apparently suitable areas suggests that considerable tolerance of crowding does exist. In summary, the shore line is a preferred pupping area, but the avoidance of overcrowded conditions induces some cows to move inland. Both habitat preference and competition for space determine the dispersion of cows; herding of cows by bulls is not an important factor.
Cows sometimes appear on the rookery weeks before giving birth. Near the time of pupping they choose a site which they occupy for up to five days before giving birth.
Interactions between cows must influence their distribution. They threaten one another for the slightest disturbance and fight for favoured rest spots such as ledges, or shady places on hot days. Cows do not bite, but confine themselves to pushing and uttering threat calls. Territorial bulls usually investigate any altercation between cows in their territories and threaten one or the other until one retreats.
Births
Cows become restless shortly before giving birth, often circling repeatedly and sniffing the rocks which their hindquarters touch. They become highly aggressive towards other cows and pups. At the commencement of labour, straining begins and the hindquarters are lifted clear of the ground; circling increases in frequency and intensity. Once the foetus starts to pass, contractions become stronger. Cows often move around when the foetus is half out, as though trying to drag the pup the rest of the way.
The time for the birth process varies from a few minutes to several hours. About 60% of births were breech presentations on the Snares Islands in 1970 and births were of similar frequency by day (mainly mornings and evenings) and night. Placentas were passed at birth in about 20% of instances, but usually came away within a few (< 6) hours.
A detailed description of the birth process and early mother-pup behaviour is given by McNab and Crawley (in press).
Post-parturient Behaviour of Mothers
Immediately after giving birth most cows are placid and tolerant of neighbours, perhaps because of fatigue. Shortly afterwards, cows with newborn pups are more aggressive than those with older pups.
Cows pick up and smell the pup repeatedly during the first 30 minutes after birth but make no attempt to remove foetal membranes where these are still enclosing the pup. Mouthing, raising and frequent rubbing of the body of the pup by the mother are indulged in for some time following birth. In many cases, cows deliberately lie on the pup and seek this type of physical contact for up to an hour. Mouthing and lifting may still occur days after birth.
Auditory communication between pup and mother is the most characteristic sound in rookeries. Pups start calling immediately after birth, but cows respond only after 30 minutes or so. Cows give a quiet lowing noise, while pups emit a ‘female-attraction call’.
Mothers retrieve their pups if they wander away. The cows try to lure pups in the appropriate direction but occasionally they pick them up by the scruff of the neck and carry them.
Schedule of Cows
Most cows spend up to five days in the birth area before giving birth, although they may have been ashore elsewhere for weeks. They come into oestrous about eight days after giving birth, and normally leave the rookery a day or two after copulation for their first feeding trip. They are normally away for three or four days at first, and on their return suckle their pups for a similar period. Thereafter, the feeding trips tend to be of longer duration than the suckling periods.
While on shore, cows spend about 60% of their time with their pups, about half of which is spent suckling. The remaining time is spent wandering around the rookery or lying on some favoured ledge away from the pup.
Newly-arrived cows are wet and apparently attractive to pups. On arrival, some cows immediately begin calling and move to where they left their pup, while others do not call but move directly to the correct locality and accept their pup after smelling it. Normally, both pups and cows shout on approach. Cows threaten pups until they
Female-Female Threats
Threat and submissive behaviour are similar in cows and bulls, but cows are less likely to attack one another, and show greater flexibility in postures and sequences. Larger cows usually win any squabbles, but cows with newborn pups are the most aggressive. The disturbance caused by cows moving about the rookery leads to most interactions, but competition for cool places and favoured resting ledges is important.
Females and Other Classes
When bulls approach cows it is normally to herd or investigate them, and cows resist this attention with open-mouth threats, snorts and growls, and often make aggressive jabs at the face. They show submissive behaviour if a bull is aggressive. At all times other than during copulation, cows try to keep their hindquarters away from bulls. Large subadult males dominate cows but all others are subordinate to cows.
Cows threaten foreign pups by intention movements rather than by threat postures and vocalisations. They show most aggression towards foreign pups which they displace from resting spots.
Pups
The black pups are born during the hot summer and form part of a densely packed herd. Each pup is protected by its mother for about ten days, but thereafter must fend for itself for long periods among a crowd of indifferent, even hostile, adults. Maternal solicitude is evident when the pup is very young but decreases rapidly. The pup is mainly responsible for maintaining the mother-pup relationship. The cow returns to the birthplace to locate her pup after an absence at sea and it is up to the pup to be there. Strong associations do not persist after weaning, which may take up to a year.
As pups become the dominant class numerically life becomes less hazardous for them. Early in the season, pups escape injury from trampling only by hiding in crannies or behind rocks, or by assembling in open spaces.
Pups form loosely-knit pods, usually of four or five animals, after a couple of weeks, when sufficient of the mothers are away on feeding trips. The aggregations allow mutual protection and companionship
Pups play with one another, indulging in mock battles. They tend to be fairly quiet as a group, but engage in constant movement. They sleep very soundly, with the hind flippers folded forward under the belly and the fore flippers folded back between the hind. The neck is stretched out and the head rested on the rocks. Some pups curl up alone while others prefer to snuggle together.
Pups moult at about two months, the moult taking up to four weeks, and are thereafter termed yearlings. They are still confined to the rookery and continue to suckle for several months.
Subadults
Subadults are noisy and active. They are numerous at the rookery during the winter and also predominate in the winter hauling grounds around the coasts further north. During the breeding period some of the subadults occupy offshore stacks and rocky areas on the fringes of the main rookery. Females mature earlier than males and large subadults are therefore almost certainly males.
The ratio of breeding bulls to cows in breeding colonies is difficult to determine. As pupping places coincide with territories previously formed by bulls, there is a grouping of cows around single bulls which gives the impression of organised groups or ‘harems’. However, until cows are restricted to one spot by having given birth they are able to pass relatively freely between territories despite all efforts of bulls to prevent them. Before the peak of pupping there is therefore continual change in the distribution of cows and new ones are continually arriving. Following the commencement of pupping, many cows are confined to a territory because they must remain with their pup at the birth place for the first ten days or so, but at any one time there is an unknown number of cows away on their first post-partum feeding trip.
Because of the continual shift in harem populations accurate counts of cows are difficult, and harem size is usually calculated from the number of pups eventually born in a particular territory. Clearly, the over-all number of cows using a particular territory cannot be determined from the number present on any one day, which may be quite small and is variable. Essentially, the territories cope with a passing parade of cows, and the tempo of pupping bears a direct relationship to the number of cows in territories at a particular time.
While taking the above into account, it is possible to calculate some measure of the bull : cow ratio by relating the total number of territorial bulls and breeding females in the colony on several days. At the peak of the breeding period on the Snares Islands in 1970, the maximum recorded ratio of territorial bulls to cows was 1 : 7.4 and the mimimum was 1 : 2.5. However, as the cows were not distributed evenly the number of cows per territory was variable: an average grouping in a single territory was 4 and the maximum observed was 12 (Crawley, 1972). It is possible also to determine harem sizes from counts of pups; pups do not move away to sea, and each cow bears only one pup, allowing a direct conversion of pup to cow numbers. On the debit side is the fact that the number of pups present at any specific time is difficult to assess, because they lay under or behind rocks, in cracks, and behind adult seals. Using this method, the average ratio of bulls : pups on the Snares was 1 : 5 and on Open Bay Islands in 1970-71 was 1 : 6.1 (Miller, 1971).
This account of the natural history and behaviour of the New Zealand fur seal summarises present knowledge of the subject but also, and perhaps more importantly, reveals areas of ignorance. The topics covered most fully are distribution, abundance, habitat use, breeding biology and behaviour, while foods and movements are given only a sketchy treatment. Conspicuous by its absence is the important subject of population dynamics. The coverage described reflects the availability of reliable information on the topics concerned, and it is clear that the dynamics of the fur seal populations should now be given priority in allocating research time and money.
Knowledge of population dynamics is essential for the planning of programmes for the conservation, management or exploitation of the fur seal. To gain this knowledge, it will be necessary to obtain data on the sex and age structure of populations, natality and mortality rates, rates of growth and development of young, productivity, and numerous other population parameters. This will involve further development of techniques for marking animals and determining their age and reproductive condition, and improvement of existing census methods. Some of this work is already being undertaken by members of the Zoology Department, University of Canterbury, with the aid of finance from the Fisheries Research Division of the Ministry of Agriculture and Fisheries.
We wish to thank the many fishermen who so kindly provided transport to and from islands and remote parts of the New Zealand coast. Financial aid was provided by the Fisheries Research Division.
When Cockayne (1928: 408) suggested that ‘New Zealand possessed a flora, part of which had originated on her own soil,’ he included genera with the majority of their species endemic to New Zealand but with one or more species occurring elsewhere, in Australia, South America, or on Pacific islands. He assumed this element to be of New Zealand origin and to be ancient (by which he meant Tertiary), called it ‘Paleozelandic’ and implied that some of its members had migrated from their place of origin in New Zealand to Australia, Malaya or South America.
It cannot always be concluded that a genus originated in the area that now supports the greatest number of species; nor should we include taxa (like Phyllocladus) that now apparently occupy only a part of their former range (Couper, 1960a). There remain, however, a number of taxa of animals as well as of plants, for which a New Zealand origin seems extremely probable, although this conclusion can seldom be tested by paleontological evidence.
The presence of characteristic New Zealand organisms in the adjacent islands that rise from the system of submarine rises to the north-west, east and south has been taken for granted, without much consideration of its implications.
Judged by recent paleogeographic studies (Fleming, 1960; Stevens, 1974) most of New Zealand's outlying islands have not been connected with the mainland during the Tertiary. Even those over continental crust such as the Chatham, Auckland and Campbell islands have not necessarily had a continuous history as land but have mainly been established by Tertiary volcanism on up-domed portions of the sea floor. Moreover, islands in the sub-antarctic zone suffered severe cold climates (if not glaciation) during the Quaternary ice ages, so that many of their inhabitants, both plant and animal, are young colonists from across the seas.
The youth of many colonists on offshore islands is shown by their close relationship to parent stocks in New Zealand. Thus the New Zealand Nestor parrots formerly living in Norfolk and Chatham
Prosthemadera honeyeaters at the Kermadec, Chatham and Auckland islands, and Hemiphaga pigeons at Norfolk and Chatham islands range from populations identical with the mainland
(Nestor productus) a moderately well-differentiated species (data from Matthews, 1931). The Auckland Island Prosthemadera must be a post-glacial colonist of the last 10,000 years and there is little reason to assume much greater antiquity for the others. The presence of the New Zealand monocotyledon Phormium tenax (Agavaceae), otherwise generically endemic, at Norfolk, the Chatham and Auckland islands and of the palm
Endemic genera of cicadas evolved in New Zealand, apparently from Australian colonist stocks, probably during the Tertiary, and their adaptive radiation, to occupy stations from coastal sand-dune to mountain fellfield, from rain forest to bare river-bed, was conditioned by the diversification of New Zealand environments during the Quaternary. Members of two species groups in the genus Kikihia Dugdale, showing a speciation pattern no older than Quaternary, have colonised islands, the muta group occupying Norfolk Island and the Chathams and a characteristic green cicada (K. cutora) reaching the Kermadec Islands (Fleming, 1973). Their New Zealand affinities stand in sharp contrast with those of Lord Howe Island cicadas which were unambiguously derived from Australia, and those of Fiji and Samoa which were derived from the Oriental region (Fig. 2).
The random nature of such trans-oceanic colonisations or of their success is emphasised by the occurrence of a single species of broom, Carmichaelia (otherwise endemic to New Zealand with 38 species there in 8 subgenera) on Lord Howe Island and their absence from closer islands (Fig. 3).
Collospermum (Liliacea), with two New Zealand species, has single species in montane Fiji and Samoa, generally considered derivative (Fig. 3), but we may here see only relict parts of its former total range and its evolution in New Zealand may be judged uncertain (cf. Skottsberg, 1937).
The four New Zealand species of Allan (1961) is the source for distributional data on Dicotyledons.Fuchsia Skinnera and thus isolated from the South American members of the genus. Their fossil pollen record (Couper, 1960b) indicates a Tertiary history in New Zealand. There is thus strong presumptive evidence of New Zealand origin for a close relative of one New Zealand species that lives on the mountains of Tahiti, and for transoceanic transport to these oceanic volcanic islands, presumably by means of its succulent fruit (Fig. 3).
The genus Melicytus (Violacea) is endemic to New Zealand (4 species) except for one species that ranges to Norfolk Island, Fiji, Tonga and the Kermadec Islands (Fig. 3), lack of speciation in these peripheral populations strongly suggesting a recent date of colonisation. Its berry is eaten by birds.
That birds have crossed the sea from New Zealand to some of the Pacific islands so far named is evidenced by the distribution of the parakeet genus Cyanorhamphus (see Peters, 1937) which is centred on New Zealand (with 3 species) but is (or was formerly) represented on all offshore islands, some of which Antipodes Islands, Auckland Islands, Chatham Islands) have been successfully colonised twice (Fig. 4) so that they support 2 species.
The most persistent colonist, Cyanorhamphus novaezealandiae, had reached New Caledonia, Norfolk, Lord Howe, Kermadec, Chatham, Antipodies, Auckland and Macquarie islands so recently that the majority of insular populations were weak subspecies. Related species, somewhat more distinct and thus probably older, lived on two of the Society Islands when Europeans first visited them. The dates of
Even more distant regions were colonised by Hebe, which has its headquarters in New Zealand (79 species) with a few species in Tasmania, south-east Australia and New Guinea. Two New Zealand species range to southern South America, and one extends further, to the Falkland Islands. The New Zealand species show every indication of active evolution, high variability, and of incomplete speciation, so that the occurrence of two indistinguishable derivative populations in South America implies a geologically recent date of colonisation, down the West Wind Drift, probably caked to the feet or feathers of seabirds (Falla, 1960; Godley, 1967).
The beetle genera Kenodactylus Broun and Oopterus Guerin have reached Patagonia and the latter has also reached Falkland, Kuerguelen and South Georgia islands as the result of transoceanic migration of New Zealand stock from which they are derived (Johns, 1974). They thus provide a zoological parallel to the two subspecies of Hebe that have crossed the Pacific from New Zealand.
Whereas many New Zealand organisms were derived from Australia by eastward movements across the Tasman Sea, plant genera like Aciphylla, Anistome, Celmisia and Dracophyllum which are centred on New Zealand (with 20 to 58 species), but have one or a few Australian species, suggest a modest return flow. Among marine shallow-water molluscs there is an occasional example of a genus appearing in Australia after a well-documented earlier history in New Zealand.
In our own time, several species of endemic New Zealand shallow-shelf mollusca have appeared abundantly in Tasmania (Greenhill, 1965), perhaps dispersed with the assistance of merchant shipping. Dominantly westerly winds and eastward sea-surface currents are now opposed to westward transport of drifting organisms, both marine and terrestrial, but such movements may have been easier in the past if the South Equatorial Current lay further to the south during periods of Tertiary warmth than it does today.
Cockayne's opinion that a New Zealand element can be recognised in the biota of other Pacific countries is supported by the present incomplete review which includes zoological and fossil evidence not available to him. On available evidence, the geological age of colonisations by such New Zealand elements seems to have been late Cenozoic, Quaternary, or even post-glacial. The term ‘Paleozelandic’, however, is not appropriate for a group of taxa of variable Cenozoic age in New Zealand, since it implies that they are old elements (if not the oldest) in the New Zealand biota. That the colonisations discussed took place across the sea is consistently supported by their random pattern, by their apparently late date, and by the geological history of recipient islands.
This paper reports on observations and research during a visit to Chile and Argentina in 1971-72, against the background of some of the extensive literature on the ecology of Nothofagus, the southern beech, in these two countries.
The 10 species of Nothofagus in South America are variously distributed from latitude 56° S. to latitude 33° N. The southern limit is due only to lack of land, the northern limit to the aridity that accompanies the Mediterranean climate of central Chile.
In comparing South America's Nothofagus with those of New Zealand, there are several biological differences, some shared with the other temperate Nothofagus of Australia and Tasmania.
These characters are summarised on Table 1; the most striking is the lack of relation between the pollen group/genetic separation and biological behaviour. Deciduousness, not found in New Zealand Nothofagus, and vegetative regeneration, normally lacking in New Zealand Nothofagus, are indiscriminately scattered among the two genetically separated groups, named on the pollen types, ‘N. menziesii’ and ‘N. fusca’ (Cranwell, 1939) and the corresponding ‘N. obliqua’ and ‘N. dombeyi’ types of Heusser (1971).
Likewise the leaf size, small as in N. menziesii, medium as in N. fusca or large and Fagus-like as in N. moorei and N. obliqua, show no grouping corresponding to the pollen genetic groups.
The character of deciduousness, classically, is associated with climatic extremes, of cold or of drought. The Tasmanian deciduous N. gunnii, a high altitude small tree fits this hypothesis. Seven of the South American species are deciduous but show direct relation to climate only in the far south, where the evergreen species N. betuloides occupies the wetter and more isothermal western end of a precipitation gradient from west to east (Fig. 2).
By contrast, at 40°-45° S. latitude (Table 2) a relation between climate and deciduousness is at first less clear. However, Weinberger (1973) has shown by detailed climatic studies of some of the Nothofagus of these latitudes that the occurrence of deciduous Nothofagus can correspond with variations in climate. The evergreen species, N. nitida, and N. betuloides are restricted to the oceanic climate of the Cordillera de la Costa (Coastal Range). The deciduous species N. obliqua is the most thermophilic, and is found in climates of continental tendency, mostly in the Central Valley of Chile, and as
On ascending the Andes, or the inland flank of the Cordillera de la Costa, N. obliqua is replaced by another deciduous tree, N. alpina, in climates still of continental tendency. The next species upwards is the
N. dombeyi, localised in a shallow altitudinal zone in the Cordillera de la Costa, and a deeper altitudinal zone on the west and east flanks of the Andes. Weinberger's characteristation of the climate of N. dombeyi stands is of sufficient humidity and limited temperature range to be oceanic. His data is, however, only from Chile; on the eastern flank of the Andes, in Argentina, N. dombeyi forms a continuous forest out to the semi-arid and deciduous N. antarctica forest of the Patagonian plateau.
Above the evergreen N. dombeyi is the deciduous N. pumilio; it is the treeline species, in a high altitude, in a cool climate of oceanic tendency. The very plastic N. antarctica, also deciduous, forms a subalpine scrub, and occupies frost hollows, both areas characterised by Weinberger as having wide diurnal temperature ranges.
There remains the apparent anomaly of the evergreen N. dombeyi sandwiched, on both sides of the Andes, by deciduous Nothofagus; in subcontinental climates below, and humid mountain climates above in Chile; but extending into the subcontinental climate of Argentinian Patagonia.
The other deciduous species of wide distribution, N. glauca, not studied by Weinberger, has its range within the range of N. obliqua, and in similar climates. N. alessandri and N. leoni, both deciduous, are of very restricted occurrence; and in climates at the more arid range of the more common N. obliqua and N. glauca.
The dispersonal of Nothofagus in South America shows a major difference from the pattern in New Zealand. Forest type maps, both published (1964) and seen in draft, show no major gaps in the distribution of Nothofagus in South America such as are found in New Zealand. Where there is enough moisture, and summer temperature, Nothofagus will grow except on the very cloudy oceanic flanks of the Cordillera de la Costa (Weinberger, 1973). This continuous distribution in its northern range exists despite a history of vulcanism that extends into the present day, as forest destroying lava flows, and volcanic ash showers up to 60 cm deep in the last 50 years. The coppicing ability of the widely distributed N. obliqua, N. glauca, N. alpina and N. pumilio in its more northern distribution may be connected with their survival after deep ash showers.
N. pumilio, which coppices at medium latitudes (37°-48° S.) (F. Schlegel pers. comm.) but not in the south of its range to 55° S., displays a parallel to the behaviour of N. cunninghamii; with only limited coppicing in Tasmania, but vigorous regeneration by this means in Victoria (Howard, 1973).
In the southern part of its range, Nothofagus is represented by three species, only one of which reproduces vegetatively there (N. antarctica). Yet this southern range of the three species includes ice sheets today reaching sea level, and still Nothofagus has made its way on to all the glacier fragmented landscape of the Southern Chilean ‘Canales’.
Comparing this over-all distribution with the fragmentation of New Zealand's Nothofagus by Pleistocene ice, and by recent vulcanism (Wardle, 1964), one associated factor differing is the presence of terrestrial mammals indigenous to South America; and among these, notably the great variety of rodents. Little seems known of their food preferences but a parallel with the European squirrel as a seed eater, transporter and forgetful storer is apt, as one possible explanation of seed transport of Nothofagus in South America.
The isolation of the large land mass of South America from the north, from Cretaceous to Plio-Pleistocene times (Graham, 1972), has apparently been allowed sufficient genetic diversity to persist in Nothofagus on a large land mass. This diversity is expressed by the vegetative reproduction of four species and the deciduous character of six species, both characters absent from New Zealand Nothofagus. Australia, with a larger land mass, has likewise among its three species, the characters of vegetative reproduction and deciduousness.
This paper is in two parts, the first dealing with Nothofagus and the vegetation surrounding its forests in the Magellanic areas of Chile and Argentina. The second part, to appear in a subsequent issue, deals with forest distribution in the more temperate regions, around latitude 41° S., and more comparable climatically with New Zealand.
Patagonia', a patois word of the early Spanish for ‘big feet’, of the local Indians, is a name given to a region of fluid boundaries. Those Indians, now virtually extinct (Searell, 1974), moved easily on the tussock plains of the drier areas, and in the deciduous Nothofagus forests of moderate rainfall (Bridges, 1948). Westward, the dense, high rainfall forests forced the Indians to the waterways, in beech bark canoes. The early European colonists found that they, on their horses, could travel on land in the same areas as the Indians; and locally
Patagonia, in Chile and Argentina, is the country where a horse can be ridden, in tussock and open deciduous forests. Floristically, Cabrera (1958) shows that the main semi-arid vegetation types of Patagonia stretch from 32°S. at high altitudes near the Andes, reach the coast at 44° S., and then occupy land from the Andes to the Atlantic coast down to latitude 54° S. in Tierra del Fuego. A cross-section of South America, from about 40° S. to 54° S., is, topographically, a larger version of the cross-section: Westland - Southern Alps - Canterbury Plains. But in South America there is a more extensive history of vulcanism, both in the main axis and on the Patagonian Plateau. Lying, as the area does, athwart the ‘Roaring Forties’ and ‘Furious Fifties’, it experiences the same west to east moisture gradient as the New Zealand miniature counterpart; but south-easterly precipitation seems rare south of 44°. In contrast to the relatively humid New Zealand east coast plains, there are the semi-desert conditions found right to the east coast of Patagonia.
In the region of the Straits of Magellan (Fig. 1), visited by the author during research on Nothofagus antarctica in 1971-72, the climate / vegetation pattern from west to east (Fig. 2) is: Cold super-humid conditions allow only ombrogenous peat vegetation, the Magellanic Moorland, a climatic effect on vegetation compounded by the westward distribution of compact diorites (Godley, 1960). Eastward from this most extreme zone, in slightly less cloudy and hence warmer summers, forest patches of the evergreen Nothofagus betuloides appear. These forests are on more favourable sites of easily drained soil parent material within a mosaic of Sphagnum bogs on locally poorly drained sites.
With decrease in precipitation summer temperatures rise slightly and the forest becomes continuous. At about 600 mm annual precipitation forest composition changes to dominance by the deciduous, erect N. pumilio. At 400 mm precipitation this tree is in turn replaced by the multi-stemmed low N. antarctica, and around 300 mm precipitation this low forest gives way to shrubby tussock grassland of Festuca. As the chain of the Andes swings around from a north-south to west-east alignment into Tierra del Fuego, the precipitation and vegetation pattern follows the same trend.
This vegetation pattern has been known since Darwin's time from accounts of passing expeditions. Godley (1960) and di Castri (1968) present well documented recent accounts. Soil relations in the wetter zones are described by Holdgate (1961), and by O'Connor et al. (1965) in the drier zones.
The recent establishment of the Instituto de la Patagonia, at Punta Arenas, is now permitting more intensive, locally based studies. Pisano (1970, 1971, 1972, 1973) is now adding to detailed floristic and ecological knowledge of the more remote areas; his published material and detailed field knowledge contribute much to the following account.
The climatic data on Table 3 are extracted from official records for Chile (also filed at the New Zealand Meteorological Service, Agroclimatology Section, Wellington).
The west to east precipitation gradient across the Magellanic region is at lower levels than New Zealand's 7,000 mm to 400 mm from Fiordland to Central Otago, but the more southern latitude, and consequent reduced evaporation, produce similar effects on the vegetation pattern. Not shown on Table 3 is the fact that monthly precipitation is virtually regular in all regions. Summer mean temperatures show a parallel trend to the reduced cloudiness along the west-east precipitation gradient. The ranges from mean maxima of warmest month to mean minima of coldest month follow a trend from extreme oceanicity to mild continentality; annual means reflect a similar trend.
Relative humidities show a general decrease from west to east, but are still high; the moisture lack of the grassland areas is more a function of wind evaporation than of heat-induced low humidities.
Precipitation may fall as snow at any season; but local reports are that in winter snow cover is neither deep nor long-lasting at sea level. Ré (1945) shows that only about 10% of total precipitation fell as snow at Punta Arenas over a 10-year period. In the drier regions, windblown snow freezes in winter and, thawing, produces surface floods in spring and this release of moisture is an important reserve of water for plant growth in summer.
Climatic data show relationships with vegetational distribution, particularly in the west to east gradient (Fig. 2) of vegetation types. Notable also is the depression of treeline in the N. betuloides forests of higher precipitation and heavy cloud, a similar situation to that on the western flanks of New Zealand mountains. In southern South America an upward extension of woody growth above the treeline is a zone of Nothofagus antarctica as a prostrate shrub, reaching at times 100 m above the last stunted N. betuloides. This Nothofagus shrubland occupies the situation of a mixed subalpine scrub of New Zealand (Compositae, Epacridaceae, etc.). However, N. antarctica also forms a subalpine scrub zone on both flanks of the Andes, to latitude 35°, at the northern end of Nothofagus forests.
An important climatic anomaly is the presence of forest in areas of mean summer temperature of less than 10.5° C., such as Cabo San Isidro. This temperature limit is normally associated with cessation of forest growth (Wardle, 1965). Even at Punta Arenas, with a mean summer temperature of 10.8° C., the forest extends to 650 m above this station. Either the altitudinal lapse rate is very slight or the truly sub-antarctic Nothofagus pumilio is capable of erect
Apparently not only climate but geology has an effect on this vegetation: This zone, vegetated by Marsippospermum and Schoenus, Donatia, Astelia, Orebolus and Gaimardia, is indicative of poor drainage, an effect of high precipitation, low evaporation, a hydric effect aggravated by the impermeable substrate. Godley (1960) demonstrates the coincidence of the ecotone between Magellanic Moorland and evergreen Nothofagus forest with the geological contact between the compact upper Cretaceous Andean diorite and more permeable metamorphics and sedimentaries to the east. Further, he believes that the absence, in the far west, of forest, and its sparsity further inland in the Magellanic Moorland zone, is due rather to substrate effects than to temperature. However, mean summer temperatures (December-February) at Evangalistas (52° S.), where vegetation is virtually treeless, is 8.6° C., well below the 10.5° C. limit associated with forest growth (Wardle, 1965); such temperature is comparable with Campbell Island, at a similar latitude.
Lack of meteorological stations in the Straits of Magellan, eastward until Cabo San Isidro, prevents determination of climate relations along the transition from Magellanic moorland to evergreen forest. However, at Cabo San Isidro, well within this forest zone, mean summer temperature is still only 8.9° C., but mean maxima for the same period is 12.4° C. Within the transition zone Pisano (1970, 1972) has described soil/vegetation relations in detail. He finds that the patches of evergreen beech forest of Nothofagus betuloides are associated with more fractured rock, and are on steeper slopes; more compact rock and gentler slopes do not permit adequate drainage. Substrate impermeability coupled with precipitation, estimated at 1,750 mm, allow only the growth of Sphagnum peat and associated mire vegetation. Young (1972) reports from Isla Desolación (52° 31′ S., 74° 31′ W.) dwarf forest of N. betuloides, in ravines sheltered from wind. These forests appear to show a distribution related to adequate drainage and shelter on this most exposed western coast.
N. betuloides (‘coigue’ is a straight-stemmed evergreen tree, with upcurving heavy branches and a large crown when mature. It has coriaceous, small, toothed leaves up to 2 cm long.
In relation to precipitation it is analogous to N. menziesii (cf. Wardle, 1967). N. betuloides, in the Straits of Magellan area, forms continuous forest in a precipitation range from about 600 mm up to about 1,750 mm to the west of Brunswick Peninsula (72° 30′ W.). As precipitation increases, and secondarily as soil parent material becomes dominated by compact diorite, the forest becomes more fragmentary, in a mosaic of mire vegetation.
the westward are seen around the Primera Angostura (72° 15′ W.).
The distribution of N. betuloides is associated with oceanicity of climate; at Cabo San Isidro, well into the zone of continuous
N. betuloides, the temperature range is: January (mean maxima) 12.4° C. to August (mean minima) 0.1° C. By contrast the zone of the deciduous N. pumilio has a range of 15.2° C. to —0.3° C. When the two species are together, for instance north-east of Lago El Parillar at 400 m, and in the Parque Nacional Lapataia at 200-400 m, N. betuloides occupies mid-slope areas, above any extreme cold air accumulation. Pisano (1971) in the Fiordo Parry area reports occasional N. pumilio trees in N. betuloides forest, near sea level on freely drained morainic deposits; apparently here the species mixture is of edaphic control, with the moisture-demanding N. betuloides not able fully to compete against the more mesic N. pumilio.
Under slightly lower precipitation, an estimated 800-1,000 mm at Cutter Cove, Pisano (1970) reports pure N. betuloides forest to sea level; but only on steeper, well drained slopes with fractured rock as soil parent material; otherwise the vegetation is treeless.
At the wetter end of its range (Bahia Morris, Isla Capitan Aracena) Pisano (1972) estimates a precipitation of 1,750 mm. Here N. betuloides forest is established only on colluvium giving free drainage; on the compact crystalline substrate, trees are absent.
The association of N. betuloides with high precipitation and humidity is reflected in the physiognomy of the N. betuloides forests: bryophyte and Hymenophyllum cover is dense on ground and fallen logs, although not swathing the trunks as does similar epiphytic cover
N. menziesii under superhumid conditions in New Zealand. The logs and standing dead branches are ‘soggy’ in decomposition and frequently blue stained inside, as are those of N. menziesii.
A stand on a well drained site of 20° slope examined in detail by the author (January 1972) at Rio Caleta, Seno Otway, had the following composition:
The soil profile was silt, colluviated with schist stones: 2 cm litter/10 cm humus, medium reddish brown/20 cm leached sandy silt, light grey with 20% ferric mottles 20 cm silt loam, light grey with 25% ferric mottles.
The mottling by translocated iron presented a very similar profile to those developed under N. menziesii on loessic colluvia of the Tararua Mountains of New Zealand under 3,800 mm of rain at 600-800 m altitude.
N. betuloides seems tolerant of moderately poor drainage—and can grow on deep peats (> 2 m); for instance on a ‘pakihi’ type terrace seen near the stand just described. However, the scattered trees are only 1-2 m high and growing always on hummocks, where drainage is adequate. Such behaviour contrasts with the growth of N. antarctica on deep peats in the N. pumilio zone.
In precipitation of 800 mm and higher it appears that the greater cloud cover and lowered temperatures produce a lowering of treeline of N. betuloides equivalent to that on western flanks of New Zealand mountains. ‘Mountain scrub’ is described by Pisano (1970) only from one area: Cutter Cove. On well drained soils this scrub includes Embothrium coccineum, Berberis ilicifolia, Baccharis magellanica. At the other two localities, Fiordo Parry and Isla Capitan Aracena (Pisano, 1972, 1973), the forest gives way at the treeline directly to subantarctic cushion vegetation including Bolax borei and B. gummifera, Donatia fascicularis, Phyllachne uliginosa, Caltha dionaefolia, Drapetes muscosus and Astelia pumila and occasional prostrate Nothofagus antarctica.
In all three areas mentioned above Pisano records N. antarctica as a prostrate shrub above the N. betuloides treeline, but the upper
N. antarctica is specified only at Fiordo Parry, at 400 m — that is, 100 m above the uppermost N. betuloides. At Cutter Cove N. antarctica is recorded as a small tree in the ‘mountain scrub’, at the other two areas as a prostrate shrub on better drained slopes in the Bolax cushion vegetation.
Similar conditions were observed by the author above the N. pumilio treeline at 600 m on Cerro Condor (Parque Nacional Lapataia) and at 650 m on the mountains above Punta Arenas. There, on slopes over 10° the N. antarctica shrubland was dense and up to 2 m high, but was prostrate and scattered in the Bolax cushions on gentler slopes. Dense, tall N. antarctica scrub was observed from the sea on southward slopes of Peninsula Brunswick, extending over a depth of 100-150 m down to the N. betuloides forest.
Stunted, isolated stands of N. antarctica were found by Young (1972) at ‘about 500m’ on Isla Desolacion, above equally isolated patches of dwarf forest of N. betuloides. The subalpine scrub belt on N. antarctica is here fragmented by exposure to wind, and by the very limited edaphic suitability of the westernmost parts of the Magellanic area.
In comparing treeline conditions in the N. betuloides zone with those of the N. menziesii forests of wetter mountains of New Zealand it can be seen that only in one case in the Magellan Straits area has there been described a mixed shrubland equivalent to New Zealand subalpine scrub of cloudier depressed treeline conditions. In other cases N. betuloides krummholz gives way directly to the subalpine cushion vegetation, with N. antarctica as a shrub only on better drained sites.
It appears here that in the far south and right through its range to 35° S. N. antarctica is playing alone the role of the mixture of species which form subalpine scrub in New Zealand. In its relationship to the taxa of the forest below timberline the subalpine ecology of N. antarctica is close to that of Pinus uncinata var. mugho of the European Alps and the French Massif Central. The Arctic equivalent in Norway and more humid Sweden is a change from forest of Pinus silvestris to tall (2-3 m) scrub of Betula tortulosa, which in turn gives way to a lower scrub of Salix spp. of 40-60 cm high.
N. betuloides appears to reproduce exclusively by seed and according to Sra. N. Goodall, of Ushuaia (pers. comm.) has produced seed there annually in the last five years. Summer conditions there in 1968-69 and 1969-70 were apparently milder than usual, and it is possible that the trees were showing similar reaction to warmer than usual summer temperatures, as does Nothofagus in New Zealand.
N. betuloides can grow to 25-30 m height with 10-15 m of clear trunk with diameters of up to 75 cm (b.h.). It is milled in the few
N. pumilio, and apparently advantage is taken on this to market a mixture of both species. Little of the high precipitation area occupied by N. betuloides is farmed and the only exploitation of the zone is by random logging of Pilgerodendron uviferum trees in the wetter, westward area, and only in areas accessible by sea. Much of the area both of N. betuloides and of Magellanic moorland in Chile is now incorporated in national parks (Pisano, 1973).
Conifers in the Magellanic area are represented only by the tree Pilgerodendron uviferum, generally on moderately poorly drained soils, and at low altitudes in the N. betuloides forest zone. It is notably absent from the Fiordo Parry area (Pisano, 1971) which appears to have a cooler climate than other western areas described by him. In the more eastern forests of N. betuloides on Peninsula Brunswick and in N. pumilio forests of even drier and more extreme climates there are no conifers.
The prostrate Dacrydium funckii of bogs further north is also apparently absent from the Magellan Straits area and Tierra del Fuego. Only in the extreme north-west of Magallanes Province (49° S.) does Podocarpus nubigena appear (Pisano, 1973). It is found in forests dominated by N. betuloides and accompanied by Weinmannia trichosperma, Tepualia stipularis and Lomatia ferruginea. Temperature conditions are evidently more favourable than further south as the estimated 2,000-6,000 mm of rain still permits forest growth. The family combination is far more familiar to New Zealanders, and puts into contrast the combination of the familiar Nothofagus with the unfamiliar Berberis, Maytenus, etc., associants of more southern forests. Other conifers play an increasing role in Chile and Argentina, north of 48°. Godley (1960) and Schmithusen (1960) give accounts of their distribution in Chile and Cabrera (1958), of their extension into Argentina.
The diminution southwards of a coniferous component with increasing cold, and then its final disappearance eastwards in the lower rainfall forest climates is a character not found in the present New Zealand vegetation. However, certain fossil pollen floras of the penultimate glaciation in the Wellington area show very low quantities of conifers (McQueen, 1973) but abundant beech pollen; suggesting that climates of this epoch could have found their equivalent at 52° - 55° S.
Nothofagus pumilio (‘lenga’) is a straight-stemmed tree, with a nearly cylindrical trunk when mature, and a very small crown. The leaves are toothed, and up to 2 cm long and relatively soft. They appear in late November and are shed in late March.
N. pumilio fits well with generalised theories of deciduous tress being those of more extreme climates. Climate data from Punta Arenas (Table 3 show slightly wider extremes than stations in the zone o the evergreen N. betuloides; extremes amplified by the dessication of the constant, strong westerly winds in summer.
In stands on freely drained soils, and from 0-600 m altitude near the coast, and 0-700 m inland, N. pumilioforms dense pure stands of trees varying from 30 m height in sheltered positions is 15 m on exposed plateaus, with diameters (b.h.) from 1 m down to 30 cm in canopy trees. Undergrowth on a moderately leached soil at 200 m, near Lago El Parillar, Chile (200 m alt.), is sparse of:
On a moister nutrient accumulation site a herb cover is encountered, as at 200 m in the Parque Nacional Lapataia:
The soil forming characters of N. pumilio in the South Patagonia climate are indicative of an acid raw humus regime. Raw humus under N. pumilio can be up to 15 cm deep, with pH 4; field evidence of leaching and podsolisation is frequent, despite Holdgate's (1961) reservations on these processes.
Altitude and increasing humidity at higher altitude may be associated with the formation of true podsols under N. pumilio. For instance at 400 m on the Cordillera Chilena, inland from the Straits of Magellan, there was no visible iron translocation under N. pumilio on well drained soils; there were ground water podsols only in poorly drained depressions. At 600 m in the same region, a 20 cm thick iron pan had formed in a leached sand beneath N. pumilio.
Likewise, near Punta Arenas, at seal level, profiles under N. pumilio showed only yellow brown, diffuse iron in the soil, but at 400 m altitude at Lago El Parillar, podsolisation was continuous, with 10-20 cm of bleached sand overlying a 2-3 cm thick iron pan.
This second pair of soil comparisons was near the Magellan Straits; temperatures there are noticeably lower than in regions to the north, so that the treeline of N. pumilio near Punta Arenas is at 600 m, while on the Cordillera Chilena it is about 700 m. The raising of temperature zonation is thus reflected in an effect on podsolisation. But complicating the problems of podsolisation in Magellanes is the question of soil parent material. Volcanics are not mapped in recent geological maps of Magellanes, but the whole of South Patagonia is mapped as ‘under the influence of volcanic ash’ (Valdés, 1969). It is known in New Zealand that podsolisation is rapid on acidic tephra soils, so possibly in Chile the same acceleration has occurred. The usual soil parent material in eastern Magellanes Province is a sandy silt (? loess + tephra) of 0-60 cm depth, dependent on topography, overlying till, and soils formed from this combination, at lower altitudes, under N. pumilio were similar to New Zealand's yellow-brown podsolic soils.
N. pumilio is less tolerant of poor drainage than N. betuloides. Within the main zone of distribution of N. pumilio between 400 mm and 600 mm of precipitation Nothofagus antarctica is the species occupying intrazonal sites of waterlogged and peaty soils. The physical characters of a sequence of soils overlying a compact Tertiary
N. pumilio on mineral soil through to a peat of > 2 m depth, showed that, when aeration porosity of the soil of the rooting zone was below 18%, N. pumilio was replaced by N. antarctica.
In the deciduous N. pumilio forests the bryophyte-Hymenophyllaceaemat is absent, and the rate of deadwood decomposition is very slow and ‘dry’, so much so that a fire can be easly kindled under any weather condition in South Patagonia N. pumilio forests. In this respect the map of distribution of Indian tribes (Goodall, 1970) shows that the land dwellers, Tehuelche (Ona and Haush), occupied the N. pumilio and drier zones to the east; avoiding the N. betuloides forest: in such a climate these tribes were dependent on good fires. This character of ‘dry’ decomposition does not seem to be an inherent character of the wood of N. pumilio: further north at latitude 40°, near S.C. de Bariloche (Argentina), N. pumilio wood near the timberline was decomposing damp, with the blue fungal stain frequent in N. menziesii wood and N. solandri var. cliffortioides wood under higher rainfall conditions.
Despite the different leaf form and seasonality some equivalence can be found between forests of N. pumilio and N. solandri var. cliffortioides. They are both of lower rainfall areas: 600-800 mm at sea level in southern Chile and Argentina at latitude 50°-55°; and down to about 1,000 mm on normally drained (not excessively dry soils) in New Zealand. Both the South American and New Zealand trees ascend to the treeline. The forest floor under N. pumilio is frequently herbaceous, especially on nutrient accumulating lower slopes, and lacks ferns larger than Blechnum pennamarina. N. solandri var. cliffortioides forests may have a similar herbaceous floor vegetation in lower rainfall areas, especially when browsing mammals have destroyed the shrub and fern layers. Cattle, horses and guanaco inhabit most southern Chilean forests, so that the similar floor vegetation may be a common character of animal treatment.
N. pumilio regenerates well under its own shade, only by seed, but their survival seems jeopardised by bush grazing of cattle. Regeneration by seeding after logging is equally good, possibly due to the ‘creaming’ selection of only the millable trees, and the leaving of malformed trees wich act as seed trees.
Most construction of houses in Magellanic Chile is based on N. pumilio; straight logs down to 50 cm b.h. and 6 m length are felled; selecting from straightness. The sight of sawn logs leaving the bush across the tray of truck indicates the small sizes taken; but these logs could be up to 250 years old. The use of floorboards 5 cm wide also
Pinus silvestris was being encouraged in cut-over land. Much millable forest of N. pumilio is on privately owned land, as backblocks of sheep farms, and the more usual policy after felling is burning and seeding of grass on ash. Pisano (1973) points out that the lower altitude areas of N. pumilio forest, coinciding with a climate of reasonable agricultural potential, has been considerably reduced. This deforestation, often unaccompanied by milling, has resulted in log-strewn landscapes sadly reminiscent of New Zealand's attempts at clearing beyond the limits of practical agriculture. Possibly the extreme is in the Parque Nacional de Torres del Paine, formerly a sheep station. Here the approach to the Southern Ice Sheet of Patagonia is a day-long tramp through skeletons of N. pumilio, burnt to the edge of the ice sheet and up to the timberline. Few extensive areas of N. pumilio forests are included in national parks although some small forest reserves include stands of this species (Pisano, 1973).
The establishment of pasture of Dactylis glomerata on these old forest soils is at first very successful, as the burning has mobilised nutrients locked in the humus. However, fertility appears to drop rapidly and pasture invasion by Aceana spp. and Baccharis magellanica (a Raoulia-like plant) is frequent.
Fertiliser is not easily available in Chilean Patagonia; all transport from the north is by sea. Reports of experimental fertilisation with N, P or K are discouraging. Response on one farm at 400 m was obtained only from sheep daggings which encouraged grass growth and raised the pH after three years by one unit.
Pasture organic matter, as in the forest, decomposes slowly in the cold, and effectively dry, windy climate: the relative humidity at Punta Arenas is around 70% during the growing season, November-March. This lack of decomposition of organic matter is probably a factor in lack of response to mineral fertiliser.
It is a generally held opinion in Magallanes Province that it is only the cold that inhibits organic matter decomposition by micro-organisms: but forest floor litter decomposition in the wetter N. betuloides forest appears similar to that in cool, humid N. menziesii forests in New Zealand.
Ploughing in of an artificial pasture of Dactylis glomerata or of native tussock in the semi-arid zone produces the same appearance, one year after, of a straw covered piece of land, the grass bases and leaves having not decomposed.
These field observations agree with conclusions of Dr. R. Schaeffer (FAO and ORSTOM, Paris), a microbiologist who spent a considerably longer period in Chile: and also with the report of O'Connor et al., 1965, that nitrification is generally at a very low level in Magellanic soils.
Nothofagus antarctica (nirre) is a small tree reaching 15 m only, or shrub, frequently multi-stemmed, and coppices centrifugally by death of central stems. It also produces root suckers in soil conditions from the wettest to the driest. The foliage is deciduous, appearing in late November, and falling in April, in the far south. The leaves are small, < 2 cm and toothed. Flowering was observed in summer 1971-72; the preceding summer had been abnormally warm; it is not known if such flowering is regular.
This species is of wider ecological tolerance than any other South American Nothofagus. It was chosen for this reason as the subject of a detailed study of its soil and climate tolerance, and the account below is extracted from current research, using quantitative methods of subaerial growth and productivity measurement, and quantitative soil physical parameters. Godley (1960), Dimitri (1962), Clarke (1964) and Pisano (1973) are among authors who have described in general terms its ecology and distribution in its range, from sea level to 650 m at 55° S., to a 200-300 deep zone at 2,000 m, at 35° S.
It grows in southern Patagonia in four sites: 1. semi-arid regions; 2. hydromorphic soils; 3. subalpine shrubland; 4. temperature inversion basins.
Ecotypes may exist appropriate to each site; there is little if any morphological variation; Wardle reports (1971) that seed of Argentinian timberline provenance grown at sea level in New Zealand has produced persistently prostrate saplings after three years growth and (pers. comm.) is still prostrate after five years.
1. In the driest wooded zones in southern Patagonia it is the only tree in a precipitation range from about 400 mm annually (east of Punta Arenas) to about 300 mm further to the east. Auer (1951, 1966) produced evidence of recent climatic deterioration by lowered precipitation, based on heavy mortality in stands of N. antarctica in the drier wooded areas of Tierra del Fuego and Patagonia. With such a drought tolerance N. antarctica has an important role to play in conservation on the fohn-swept areas east of the Andes and their extension into Tierra del Fuego.
Nothofagus antarctica forests, in the drier regions of Patagonia are generally fragmentary in distribution, and it appears that fire and land clearing have played an important role in this distribution (Pisano, 1973). Wood charcoal was found at depths of up to 50 cm in soils in this region, and the existence of degradation sequences from forest
The vegetation pattern, in the forest-grassland mosaic occupied by N. antarctica can be related to residual soil depth. The basic parent material consists of a varying depth of friable sandy loam, overlaying a compact till; the upper sandy silt appears to hold increasing quantities of undecomposed organic matter in a moder form. The bulk density
of upper mineral soils under the semi-arid forests of N. antarctica is very low (0.5-0.7), lower than expected for the corresponding organic matter values; the presence of volcanic ash (Valdés, 1969) may account for such an anomaly.
The following sequence shows the relation of depth of friable soil to vegetation cover.
In this sequence, only the two N. antarctica sites were studied in detail and considerable differences in moisture storage were demonstrated there. Not only was the total moisture storage higher in the deeper profile, but the soil held more moisture per unit volume.
Degradation sequences of this type at 100-300 m altitude ceased at 300 m; where higher precipitation allowed growth of N. pumilio and replacement of tussock grassland by subantarctic cushion vegetation of Bolax gummifera and Azorella spp. on slopes below 5°.
2. A second type of site already documented in some edaphic detail (Holdgate, 1961) is on hydromorphic soils in areas of climate supporting forests of taller species of Nothofagus on better drained soils. Current detailed studies of soil aeration show that on a shallow peat (20 cm) radial growth rates of N. antarctica were equal to those on a well drained valley bottom site in semi-arid country. The trees on shallow peat were 3 m high; on 80 cm of peat, growth was stunted to 1 m high.
3. Also reported by several authors, including Schmithusen (1960), Dimitri (1962) and Pisano (1970, 1971, 1972) is the occurrence of N. antarctica in subalpine shrubland zone, throughout its latitudinal range. In a situation studied in detail by the present author, erect
N. pumilio forest gave way at 600 m to a narrow zone (30 m) of krummholz of the same species. At 650 m, on slopes of 10° or greater, there was a dense 1.5 m high shrubland of N. pumilio : N. antarctica at a 1 : 4 proportion in cover. The two species were growing with stems running downslope for 2-3 m, then ascending. Radial growth of N. antarctica, at 1.8 mm/year, exceeded that of N. pumilio at 1 mm/year.
On a poorly drained area of 1° slope at 650 m altitude, N. antarctica grew as scattered shrubs, 30 cm high, with radial growth of ca. 0.5 mm/yr. The accompanying cushion vegetation of Bolax gummifera, Azorella lycopodioides and Empetrum nigrum accounted for 90% of the ground cover. Here N. antarctica was showing its tolerance, not only of above treeline conditions, but of impeded drainage.
4. The last type of site is extensive, and is occupied by forests of N. antarctica in well drained valley bottoms. Such zones are extensive between Rio Rubens and Puerto Natales (it is probable that Kalela, 1941, took his ‘Rio Rubens’ growth rate samples from such sites) and to the north of Puerto Natales. The hills around carry dense N. pumilio forest, and a probable reason for the change of pattern to N. antarctica is associable with cold air drainage.
The valley bottom forests of N. antarctica include some of the highest N. antarctica (15 m high, 80 cm diameter) seen in Southern Patagonia; however, almost without exception the upper third of the trees was dead; no reason was established, but the uniformity of the kill line suggests climatic accident, such as an abnormal freezing wind in spring.
Studies at present under way indicate that N. antarctica in Southern Patagonia is a slow growing tree: the highest increment recorded by Kalela at Agua Fresca, south-west of Punta Arenas in ca. 500 mm of precipitation, is 1.18 mm/yr radial growth. In more arid areas of Maria Christina, north-east of Punta Arenas, studied by the present writer, 1.50 mm/yr was recorded in a favourable valley bottom site in ca 330 mm annual precipitation. Such growth rates are equalled in shallow peat (see above) and only exceeded above the treeline, where 1.80 mm/yr was recorded, but there height growth was very limited. The over-all effects of subantarctic cold temperatures, especially of summer, are seen in all these growth figures.
Despite the slow growth of N. antarctica its reproductive capacity, by seeding, coppicing and root suckering, contribute to a colonising aggressiveness and persistence under disturbed conditions, especially in the semi-arid zone. Add to these reproductive features a wide ecological tolerance, and it is difficult to equate its ecology with any New Zealand Nothofagus.
These features distinguish it from N. solandri. The latter will grow in lowest precipitation experienced by Nothofagus; it will occupy shallow peats, but not over 30cm (Elder, 1965). It will form a krummholz (Wardle, 1970) at treeline, but will not form a distinct scrub zone above treeline. N. solandri will tolerate, to a limited extent, cold air drainage basins, but certainly not of the extremity of those occupied by N. antarctica.
Nothofagus antarctica is used only as firewood, although sawlogs have been obtained from larger stands in Tierra del Fuego (Bridge, 1948). The wood appears durable; but almost always shows heartrot inside the first 50 growth rings.
Its main use should be as a shelter. and soil stabilising tree. All attempts at exotic establishment with Pinus, Cupressus macrocarpa, Salix and Populus are successful in the semi-arid zone only if permanent artificial windbreaks are erected.
These notes are taken partly from a visit to the sheep station, Condor Estancia, near Cabo Virgenes, in Santa Cruz Province, Argentina, close to the easternmost extent of Southern Patagonia at the Straits of Magellan. Precipitation here is about 250 mm annually and the area all within climax grassland. The frequent snow in winter is not generally deep, but freezing conditions until November imply a concentration of moisture availability at the beginning of the growing seasons.
The vegetation of the estancia, which occupies the whole southeastern corner of Argentinian Patagonia from Rio Gallegos to Punta Virgenes, shows a regional pattern, probably related to precipitation differences. To the south, along the Straits of Magellan, the shrub Chiliotrichum diffusum is more abundant in the grassland. According to Sr. E.
To the east, Empetrum nigrum is more common. This species, associated with gravelly soils further west, is in lower precipitation areas associated with shallow soils or with gravels. Its eastern occurrence at El Condor suggests greater soil degradation in pre-settlement times, probably in consequence of great frequency of fire.
The tussock grasslands, dominated by Festuca gracillima on drier ground and F. pallescens in hollows, have a floor cover of very different botanical character from New Zealand's low tussock grasslands.
For instance, Pisano (1973) lists, from Chilean tussock grassland:
It is only when one adds the cosmopolitan genera in the form of:
that any generic resemblance with New Zealand can be recognised.
Under acid conditions, and impeded drainage, invasion of these grasslands by Azorella reminds one of the closeness of the circumantarctic element.
The management of this vegetation for sheep grazing presents some features and problems very different from similar looking tussock country in New Zealand.
The Festuca tussocks are palatable at all stages, unlike Festuca novae-zelandiae; and there appears to be a higher proportion of palatable plants, especially stoloniferous grasses on the floor of the tussock vegetation. With this natural advantage, burning, as was practised in New Zealand until recently, is not used as a management technique for tussock grasslands in Argentina or Chile. In an unimproved state good tussock grassland in Southern Patagoria can carry three or four Corriedale sheep per hectare.
Pasture improvement techniques are still at a rudimentary stage; partly because of the difficulty of getting response from fertilisers, and from the persistent failure of clovers. It must be added that it was only in 1971 that I.N.T.A., the Argentinian agricultural research organisation, established its first permanent field officer at Rio Gallegos. Until then the nearest advisory office was at S.C. de Barriloche, thirty-six hours by road!
The pasture improvement technique developed by Sr. Blake consists of ploughing the tussock in and seeding with Dactylis glomerata, Phleum pratense and Lolium perenne (cocksfoot, timothy and ryegrass). Of these, cocksfoot is the most successful. It takes up to four years for a continuous grass cover to establish: consequently weed invasion is high, particularly by dandelion. Another technique used with
There is, obviously, much need for basic research on the soils of the grasslands. I say basic, because a considerable number of Patagonian farmers have contact with New Zealand to the extent of studying at Massey or Lincoln. All too often their empirical experimentation at home, based on far more temperate New Zealand conditions, has failed in Patagonia.
Further, O'Connor et al. (1965) postulate phosphate deficiency and aluminium toxicity. Argentinians believe that sulphur deficiency may be chronic in the area. R. Schaeffer (pers. comm.) believes that the whole cycle of organic matter decomoposition is very slow in cold and dry climates. But as yet very little continuing work has been carried out in Southern Patagonia to find the chemical and physical bases of soil fertility.
A final comment: considering the climatic conditions of a frozen-up winter and a three to four months growing season it is truly surprising to see that the natural grasslands can produce such large, healthy sheep. Possibly it will be better to leave the palatable tussock and enrich by oversowing of pasture grasses, with application as fertiliser of what elements are actually lacking, rather than attempting complete conversion to swards of introduced grasses.
I should like to express my gratitude to all those who helped my field work in Chile and Argentinian Tierra de Fuego: Sr. M. Martinic B, Director of the Instituto de la Patagonia; Sr. E. Pisano V., Assistant Director and Botanist at the same institute; Dr. F. Schlegel S., Director of the Instituto de la Silvicultura y Reafforestacion, Universidad Austral de Chile; Sra. N. R. P. de Goodall of Ushuaia; Dr. D. M. Moore of the University of Reading; M. J. Pigier of Punta Arenas and the Schlumberger Geophysical Company, both in Chile and Argentina; Sr. H. MacLeay of Punta Arenas; Sr.
This account forms part of a research project supported by:
Victoria University of Wellington Refresher Leave Grant, and Internal Research Committee Grant
New Zealand University Grants Committee, Research Grant Bourse de Stage du Gouvernment Francais
Plant names, with families, mentioned in text (excluding Nothofagus, Table 1), indigenous to South America or New Zealand.
Published by the University Press of Hawaii, Honolulu, 1973. $US18.50
This volume, the first of three proposed (Vol. 2 Brachycera, Vol. 3 Cyclorrhapha), is a solid book, sewn and bound in a hard cover. It is built to stand up to the constant use it will receive from the serious diptera student. Catalogues do not need to be explained or excused. The labour-saving devices they are is well understood by any biologist studying particular groups of organisms.
This excellent one is produced on the pattern of the successful U.S.D.A. Catalog of the Diptera in that it is a compromise between a checklist (only a list of taxa) and a complete work (references to all occurrences of every name). Thus each species entry has information on the distribution by countries, reference to the original description, the type locality. Higher categories are adequately recorded with original reference, type species synonymies and occasionally very brief additional information perhaps of a very significant additional reference or information on habits. or habitats.
No record is included on the depository of type specimen(s) and I find this an annoying ‘lapsus’. Dr. Miller put this information in his catalogue of New Zealand Diptera by means of a number code and some similar system could have been used here to advantage.
The 618 pages are needed to record 382 valid genera and 6226 valid species distributed among 23 families. Its effective dates are up to and including 1970.
This and subsequent volumes are essential references for the serious diptera worker but only the chief biological libraries in New Zealand should aim to obtain copies.
This beautifully illustrated book contains 228 paintings of native trees and shrubs, both conifers and flowering plants, with examples of all genera. Well, actually, not quite that many trees and shrubs, as one herb (Hibiscus trionum) and some climbers (e.g. some Clematis and Muehlenbeckia species, Calystegia, Ipomoea and the passion flower Tetrapathaea are included). Hibiscus trionum was included to enable comparison with the ‘shrub-like perennial’ Hibiscus
diversifolius. There are about 565 species of native trees and shrubs in our flora and therefore this book deals with approximately 40% of them. It is stated on the inside of the dust jacket that the ‘primary purpose of this book is to provide a means of identification of our native trees and shrubs’. If this were the only purpose, one would be better served by consulting Trees and Shrubs of New Zealand by Poole and Adams (N.Z. Govt. Printer, 1963). For a mere $2.50 that excellent book gives a brief diagnostic account of all our trees and shrubs and salient features of over 400 of them are illustrated by accurate and artistic pen and ink sketches. However, as is also stated on the dust cover, the paintings in Mrs. Eagle's book have a great intrinsic beauty and are the result of twenty years' work. The author is a keen member of the Royal Forest and Bird Protection Society and the book is the result of her wish to share her love of painting and the New Zealand bush and to persuade New Zealanders to ‘ease up on the chainsaws and matches’. The paintings make identification of most of the plants illustrated comparatively easy.
With few exceptions, a page is devoted to each species. Written information is kept to a minimum on these pages-there is a consecutive number for each species and its scientific names, and if the plant has them Maori or common names are given. Male and female signs (referred to as ‘Zoological’ symbols on page 240!) unobtrusively identify male and female flowers and ‘juv.’ indicates juvenile leaves. In general, each species is illustrated by foliage (life-size) with flowers and sometimes fruits attached. Enlarged paintings of comparatively small flowers and fruits are frequently included. The order of families and genera follows that in Volumes 1 and 2 of Flora of New Zealand (N.Z. Govt. Printer, 1961 and 1970). Paintings on each page are admirably uncluttered. It does seem a pity, though, that small habit paintings were not included to show whole trees and shrubs where these have a characteristic form, e.g. rewarewa ( Knightia excelsa). Only nikau palm, cabbage trees, flax and kiekie are illustrated in this way.
It is good to see some of the rarest plants illustrated - Homolanthus polyandrus and Boehmeria dealbata confined to the Kermadec Islands;
Many paintings are superb. I was particularly impressed, for example, with those of the Dacrydium species, Fuchsias, Aristotelia, broomes and cabbage trees. It is difficult to illustrate a white flower on white paper and some illustrators overcome this by showing such flowers against a background of leaves, etc. In a few illustrations, particularly of enlarged flowers, the outlines of white petals seem too
The illustrations are followed by maps of New Zealand and outlying islands and sixty pages of botanical notes which describe the plants illustrated. Mrs. Eagle notes that a substantial proportion of this information is taken from Volume 1 of the Flora of New Zealand. The description of each plant is preceded by a number corresponding to the plate number. A bolder type for these numbers would have permitted a more rapid correlation between an illustration and its text. The description includes the locality and time of collection of the plants illustrated. It would have been helpful if more information had been included for some of the illustrations. For example, anyone comparing Hall's totara and totara would be misled by the illustrations of male cones, in that the greenish-white compact cone shown for Hall's totara is not fully mature, whereas the more elongated brown one shown for totara is mature and has probably shed its pollen. Unisexual flowers are a common feature of many New Zealand trees and shrubs and frequently the flowers of one sex contain sterile organs of the other sex. For example the male and female flowers of titoki (plate 136) contain respectively sterile carpels and sterile stamens. Such features are not mentioned in the botanical notes. A useful feature of the notes is an explanation of the meaning of the family, generic and specific names, but translations of the Maori names are not given.
The botanical notes are followed by a glossary of scientific terms, a bibliography and an index. There are some inconsistencies in the bibliography - dates of publication are given for some books, but not for others. The bibliography is not extensive and it is unfortunate that some books are omitted which would be useful to those seeking further information on native trees and shrubs, including their cultivation. Thus The Cultivation of New Zealand Trees and Shrubs by L. J. Metcalf (Reed, 1972) and Gardening with New Zealand Plants, Shrubs and Trees by Muriel E. Fisher, E. Satchell and Janet M. Watkins (Collins, revised edition, 1975) are not cited.
The book seems remarkably free from errors. I can find no misidentifications of any of the plants illustrated. A minor error in the index (page 309) is the italicising of the word ‘species’ after Neopanax.
I have left the most unpalatable feature of the book until last. Its price! From the point of view of size, binding, paper quality and quality of the colour printing Eagle's Trees and Shrubs of New Zealand in Colour would seem to be more or less on a par with Mark and Adam's excellent New Zealand Alpine Plants (Reed, 1973).
Yet the latter retails for $19.50 (cloth bound) and $13.50 (paper bound). Have printing costs accelerated this much in little over a year? Perhaps Collins will consider issuing a cheaper paper-bound edition?
Dear Sir,
re: Painted Lady Butterfly
I was most interested to read the letter from Mr. Tuatara 21: 129.Cynthia kershawi) is a regular migrant to New Zealand and is found most years on the west coast of the North Island, usually during early October. Sometimes it arrives in large numbers in such areas as Taranaki and North Auckland and it is amazing to see how fresh the specimens are even after their long journey across the Tasman.
These immigrant butterflies then breed in New Zealand to produce a second generation round about Christmas time, or in January, and it is specimens of this generation that Mr. Rickard seems to have found in Hawkes Bay.
It is strange why no specimens of this species have been found overwintering in New Zealand unless they attempt a return journey to Australia, just as the Painted Lady Butterfly of the northern hemisphere migrates back to Africa with the onset of autumn.