The Pa Tuna, or Eel-weir.

Eel-weirs represented a permanent institution in many districts, as in rivers and streams, also at the outlets of lagoons and lakes. In certain districts they were not used—in some cases because eels were lacking, as in the Taupo and Ruatahuna districts. In some lakes and rivers, or parts of such, the waters were too deep to allow of weirs being constructed, and in such places eel-pots of the tarino and waharua types were used, being set in the open waters. In such cases, of course, some form of bait was placed in eel-pots in order to attract the eels, but no bait was necessary in the case of pots set at a weir for the purpose of taking the eels moving down to the sea in page 132autumn. There were two forms of weirs in former times—the V (see fig. 36) or double V (as in fig. 38), and the straight single-fence weir employed in the Whanganui Valley, but, so far as I am aware, in no other district. The former kind is usually termed a pa tauremu, and I have heard the term pa auroa applied to the Whanganui form. (See figs. 40, 41, 44-48).

The weir composed of a single fence-like barrier, as constructed by the Whanganui natives, does not extend across a river, but is erected almost in a line with the current. An offset post at its down-stream extremity replaces the other long wing fence of the V weir. This local usage was, we are informed, adopted on account of the great quantities of driftwood brought down by freshets and floods in the Whanganui River, and which would sweep away any obstruction, such as a weir, placed across the current. (See figs. 46, 48.)

Pa rauiri (and pa rauwiri) is another name applied to pa tuna, or eel-weirs; and a weir is also alluded to simply as a rauwiri. This is page 133 a descriptive name, rauwiri being a term denoting the wattled type of fence so frequently employed in the construction of fish-weirs. Weirs are constructed at rapids, where the rushing waters glide from one calm reach to another; here the waters are shoal enough to allow of fences being erected. Should the waters of a river flow down several channels at the place selected, then the principal channel is selected, the one down which eels are more likely to travel. This is sometimes ascertained by casting pieces of dry driftwood into the river some distance above the ripple and then noting the channel to which most of the floating driftwood is attracted. That particular channel would be selected as a site for the weir.

The fences of a weir were formed by driving stout stakes (matia) into the bed of the river and then converting these rows of stakes into a barrier by means of wattling long lengths of pliant manuka brush between them, thus forming a wattled fence, generally alluded to by us as a brush fence. Where suitable brush was not obtainable, then the common bracken was employed for the purpose. The packed brush allowed much water to flow through it, but when eels encountered the oblique obstruction in their down-stream course they seldom forced their way through the brush, but simply followed the line of fence down to its lower end, often to their own undoing. Occasionally, we are told, such stringy bark as that of the totara tree was used in place of bush. Manuka and bracken-fern are not found at all places hence the best material available was employed.

The term tauremu applied to the V-shaped weirs is apparently a descriptive one. The word remu denotes the lower end of anything, and the lower end of this form of weir is its important part, where the fish-traps are set. A weir having two ngutu, or outlets, is termed by some a tararua.

In small streams, and some others where no destructive floods occurred, a weir-fence might consist merely of the vertical stakes driven firmly into the stream-bed, and the wattled brush, no connecting-rail, struts, or scour-mats being employed. In many cases, however, such aids to stability were necessary, as the weirs were so frequently erected at rapids, where the force of flowing water is much accelerated, especially during a freshet, and where any obstruction is liable to cause scouring of the stream-bed.

When erecting a fish-weir the Maori generally uses stakes of manuka. The species known as white manuka provides good straight stakes in its sapling stage of growth. These stakes are pointed at one end and driven down into the stream-bed by means of striking them with a heavy wooden club, called a ta, used as a beetle or maul page 134 (fig. 37). A piece of heavy, cross-grained hardwood was utilized as a ta, the shaft or hand-grip being worked down to a suitable thickness, while the outer end was left wide, thick, and heavy, to give the necessary weight. One face of it was flattened to serve as a striking-face, and by long use a hole would be worn in the middle of this face. I am not sure that a hollow was not formed in the face of a new ta prior to the implement being used. I have known this to be done. The roots of the maire tree form excellent material wherefrom to fashion a ta. In late times the iron-ringed maul of the pakeha has been much used in place of the old implement.

Having driven the line of heavy stakes, each from 3½ in. to 5 in. in diameter, the barrier is completed by means of wattling the brush in between the vertical stakes—that is, to rauwiri the fence. Long poles some 3½ in, to 4 in. in diameter are lashed in a horizontal position well up toward the upper ends of the stakes, and on the downstream side thereof, to serve as a containing and bracing rail; this is the huahua. It may or may not be considered necessary to page 135 attach another rail lower down. Stout stakes are then inserted as struts on the down-stream side of the fence, the lower ends thereof being punched into the river-bed, while their upper ends are secured to the upper rail. All such attachments are made by means of lashing with stout, strong, and pliable aka (stems of climbing and creeping plants of several species). When aka-pirita, or supplejack (Rhipo-gonum scandens), is used for the purpose the vines are split down the middle. It will be seen later that in cases where the thrust of turbulent flood-waters has to be reckoned with very heavy, well-braced structures are erected.

The diagram given (see fig. 38) shows the form of a V-shaped weir with two ngutu, or small openings. Following the figures in order, the two wing fences of the weir extending down-stream from either bank are marked 1. These are termed paihau, pakau, and pakipaki, and act as guiding-barriers to divert Tuna from his direct course down-stream and lead him by pleasant ways into the seducing eel-pot. page 136The down-stream ends of these two wings leave an intermediate and wide space of free water in the middle of the stream. This space is then almost entirely blocked by the tuki or haumi (2), or central barrier, consisting of two similar fences that converge to a point up-stream and also act as leads to the two ngutu (4), or narrow outlets, where the eel-pots are set. Just below each narrow outlet are inserted two heavy stakes (3), known as tapangutu; pou reinga, and pou rerenga, to which is secured a funnel-shaped leading-net (7), called purangi, poha, and rohe. The small lower end of this guiding-net is inserted into the entrance of the eel-pot (8), into which the eels pass in their down-stream glide. In some cases the purangi funnel-net is secured to the posts (3) by means of temporary ties; in others the stout vine hoop that distends the mouth of the net is attached to a square frame, as shown by Mr. Downes in his excellent paper on eels and eel-weirs (Transactions of the New Zealand Institute, vol. 50, pp. 310-312). This frame is slipped down on the up-stream side of the two pou reinga posts, against which its outer parts rest, and are hard pressed by the flowing waters. The attached funnel-net and the eel-pot float free below the posts. (See fig. 39, p. 137.)

In places where scouring of the river-bed is apt to occur a scour-mat is laid down, as shown by the darkened areas marked 6. It is often necessary to extend this scour-mat well to the down-stream side of the weir. This preventive of erosive scouring is known as the whakareinga, whakatakapau, whakatahuna, and whariki. It is formed by placing a layer of brush or fern (bracken) on the river-bed and securing it there by means of containing-rods firmly pegged down. A detailed account of the procedure appears in the description of lamprey-weirs to be given anon.

The two pou reinga, or stout posts, at the ngutu, or narrow outlet of a weir, seem to have been viewed as the most important parts of a fish-weir. These were the posts that were sometimes embellished with a carved design, and they were viewed as being of a permanent nature, although the fences would need to be renovated, or possibly reconstructed, each season. It was at the base of one of these posts that the mauri, a talismanic agent, of which more anon, was usually deposited.

Mr. Downes describes and illustrates a single outlet V-shaped weir at Moumahaki in which from the down-stream ends of the two converging wing fences other lines of fencing extend to either bank at right angles to the current. The actual use of these barriers is not clear, nor do I know how far spread was the usage; it certainly was not universally practised. His description of lashing small bundles of long fronds of bracken to the vertical stakes of the weir-fence is clear, page 137as also is that of wattling manuka brush in the Stakes above the ordinary water-level. The fern is said to be more durable in water than manuka, and the writer tells us that the fern layers were trampled down into a compact mass, which was further secured by means of driving an extra series of stakes through them. The weir shown in the illustration lends itself to the fence straight across stream and to page 138 the braces described as extending from wings to shore; but it is evidently a small stream, and can scarcely be "between 30 ft. and 40 ft. wide," as stated at p. 307, Transactions of the New Zealand Institute, vol. 50. Neither do the banks "rise sheer out of the water." The description given at p. 307 does not apply to the weir shown in Plate 23.

Waiapu natives informed me that in some cases the larger weirs had a footway along the top of the fences for the convenience of persons engaged in attending to the eel-pots or nets. In the larger streams a rope or vine was sometimes secured to the down-stream side of the weir-fence for the benefit of such attendants when they made their visits to the weir in canoes. Thus a person in a canoe would haul himself and craft along the barrier from one outlet to another, by means of the aforesaid convenience. In small streams a one-outlet weir was, of course, sufficient, but in rivers they were provided with two, and sometimes even more, ngutu. These eel-weirs, be it understood, were erected for the purpose of taking eels that were coming down-stream on their way to the unknown spawning-grounds in the depths of the ocean, wherever they may be. As the Awa folk of the Bay of Plenty express it, eels go to their page 139 ancestress Wainui (personified form of the ocean), there to give birth to their young. Inasmuch as eels are descended from Te Ihorangi, and connected with Para-whenua-mea, Hine-moana, and Tangaroa, and, moreover, have their relatives Koiro, Tuere, and Para dwelling in the ocean, then assuredly they are within their rights in adjourning to the realm of Hine-moana, the Ocean Maid, in order to bring forth their young. So reasoneth the Maori.

When, in passing down-stream, Tuna comes into contact with one of the oblique fences of a weir, he naturally follows the line of least resistance, and so passes down that fence and on into the funnel-net and the attached eel-pot. But here is the application of the old-time saying. "Tangaroa ara rau" ("Tangaroa of the many ways"), for be the weir-fences never so closely wattled and the weirs on a river never so numerous, yet great numbers of eels always reach the lower-most weir, and, moreover, yet other myriads pass scatheless out to sea.

The larger or more important eel-weirs had special names assigned to them, as distinct from the names of the places they were situated at. Thus Te Ruruanga and Nga Purua are the names of weirs at Hiruharama, Te Arero is at Koriniti, while Te Rua-tangata and Te Rere are two others at Tawhiti-nui—all on the Whanganui River. Other weir-names of that river are Kaituri, Manurua, Kairakau, and Whakaurunga, together with others too numerous to mention. I have been told that eel-weirs were sometimes named after ancestors.

Whanganui natives were wont to allude to eel-weirs as tawhiti awaawa. The tawhiti is a form of spring trap formerly employed in the taking of the native rat, a prized food-supply, and the name is in this case applied to a fish-weir in a semi-humorous manner. An old local saying was, "Ka piko tawhiti karawa, ka mawhiti tawhiti awaawa" Tawhiti karawa is a term applied to rat-traps. Whether karawa is a form of karawao (the name of the supplejack among Whanganui natives) or the karawa of bird-snarers I cannot say. Possibly the saying might be rendered as "The rat-trap is set, the eel-trap is sprung"; but the meaning is doubtful. The saying is reported to have originated with one Pungatere, an ancestor of the Pamoana folk.

Some eels are taken during a flood, or, rather, we are told, during the first phase of a flood, the freshet that brings discoloured waters and is termed wai tuhi. This is the time at which the matamoe eel is taken. But the later stages of a flood are barren as to fish, and none will then be taken. These secondary flood-waters are termed tawawarua. Such spates occur about March on the east coast of the North Island, and eel-weirs are, or were, then attended to in order to take the matamoe. Should a person delay in setting his eel-trap he would miss the descending eels, and one would say, "Nau i whaka-takoto i to hinaki i te wai tawawarua, anei te wai tuhi ka taha" ("You set your eel-pot during the main flood, after the first freshet had passed"). This item was contributed by Waiapu natives.

Naturally, other fish are not infrequently taken in a trap set for eels at a weir—such species as kokopu, panako, and upokororo.

Hakaraia Pahewa, of Te Kaha, Bay of Plenty, has contributed a few notes on eel-weirs, and also some explanation of how nets were used at weirs in place of eel-pots. (See Appendix 4.) The weir he describes is the ordinary V-shaped one. He states that experts were page 141called upon to decide as to the location of an eel-weir; that the two pakau, or wing fences, constitute a pa tauremu, but that in large streams (rivers) the V-shaped weir might be duplicated, or three, even four, might be erected. A weir seems to have been described by the number of ngutu in it, the narrow opening at the base of the V where net or eel-pot was set. A four-ngutu weir was probably not often erected. My informant, it will be observed, applies the term eia to the current of a stream, a curious variant of ia; while another contributor, of the Matatua district, applies the term awai or a wai to the same. Hakaraia explains that the first process carried out in erecting a weir was the fixing of the lines of stakes (pou and matia), which were driven into the river-bed about 18 in. apart, though the distance was regulated according to the length of the manuka brush to be used as wattling-material. He applies the term takaka manuka to this material, probably because the slim stems of the manuka, of about 1 in. in thickness, were utilized as well as the leafy brush of the upper parts. Takaka has apparently such a meaning as has kaka, which denotes a stalk—at least, in the Bay of Plenty area. He then tells us that two very stout posts were inserted at the ngutu, or narrow outlet, and that these were braced by means of fixing struts on the down-stream side, so as to enable them to withstand the rush of the water combined with the drag of a heavy take of eels in the net. Where he first makes use of the word kurupae it is apparently applied to the braces, which are usually described as hokai or whakawhirinaki. Kurupae is usually applied to beams, &c., in a horizontal position. Below the ngutu a scour-mat was laid down, and this also was composed of takaka, presumably the long slim growths of manuka, with its branchlets and leaves, though the stems (stipes) of the common fern (bracken) are known by the same name, and this fern was often utilized as a material for scour-mats, and even for the rauwiri fences of a weir. The brush mat was confined by means of poles laid across it and pinned down.

Two roau, or rails, were lashed in a horizontal position to the two posts at the ngutu, one at the bottom and one some distance higher, and to these rails were attached the hoop that distended the entrance of the funnel-shaped net. This net (rohe) was of a size that would contain many eels—one thousand, as one of my informants jauntily put it. The net was placed in the water on the up-stream side of the two posts, and the hoop distending the mouth thereof was pressed down through the water until its lowermost and uppermost parts just came against the up-stream sides of the two horizontal rails. These rails took the strain of the net, the tou, or small lower end, of which floated free down-stream. The lower end of this funnel-net was made page 142 page 143 open, but when set at a weir it was lashed tightly to prevent the escape of eels caught. It could, however, be quickly untied when eels were to be emptied out of the net. (See figs. 42, 43, 43A.)

When the potless net was in position it would be visited somewhat frequently at certain times, as when a slight spate was bringing driftwood and brush down-stream, which debris might block the narrow ngutu and injure the net. Also the attendant would grope down underwater to feel if there were any eels in the net; this would be done each time he visited the net. Should a considerable number of eels be found entrapped, then the fishermen tied the funnel-net tightly above the mass of eels collected at its narrow lower end, lifted the latter, untied it, and emptied the eels into baskets to be taken away. The tou, or small end of the net, was then retied, the page 144other tie released, and the net returned to its original position to serve as a kawau moe roa, a term applied to a set-net.

The eel known as matamoe is taken during the first freshet that comes down the rivers during the autumn months. This occurs in March or April. The smaller eels of this species come down first, and the papawhenua, or full-grown ones, are the last to appear, and such appearance betokened the final eel-run of the season. There are four different names applied to this species, apart from the generic term of matamoe:—

Rara	..	The very small eels of the species are so termed.
Oue. .	..	Applied to those of larger size.
Aopori	..	Denotes still larger ones.
Papawhenua	..	The full-grown matamoe.

Apparently these are not names of different varieties, but of the eel at various stages of its growth. Such were Hakaraia's comments on the pa tauremu, or V-shaped weir, styled a taraua by Whanganui natives.

In a later communication my correspondent speaks of takaka rarauhe and rauwiri manuka in describing these wattling-materials, and explains that the latter was used when constructing large weirs in rivers, and the fern for small weirs in streams. This seems to be the genuine usage—takaka for the bracken or fern, and rauwiri for manuka brush. Also he states that the two posts at the outlet are called tapangutu; that the kurupae are tokai or hokai (struts or braces); that the scour-mat tends to increase the swiftness of the current and to prevent eels making back up-stream and so out of the net. The hoops of the net were of supplejack or manuka; its wide end carried a somewhat wide mesh, and the narrow, down-stream end a fine mesh. Similar nets are used to lead eels into an eel-pot. This latter combination of funnel-net and eel-pot is divided by Hakaraia into six parts—i.e., six different names are applied to different parts. The wide up-stream end of the net (its waha, or mouth) is called the potaka; the second whanga, or section, is the pararakau; the third whanga is the tatao; the fourth is the kupenga-a-kai; the fifth is the pu hinaki; while the sixth whanga is the actual eel-pot.

No early writer on the Maori has put on record a good illustration of the native fish-weirs. The representation of an eel-weir in Hoch-stetter's work is a poor thing, and that included in White's volume of illustrations merely shows the two posts called pou reinga, without any wing fences to control the movements of the eels. To set an eel-pot in such a place would be equivalent to setting it in open waters. The two posts referred to in White's illustration show each a carved human face on its upper part. The weir depicted by Angas represents page 145a form of weir never seen by the present writer. (See fig. 40, p. 138.)

The following recital is a charm that was employed in order to cause eels to enter the net or eel-pot set at a pa tuna, or eel-weir:—

Te ika i Heretaunga, te ika i Ngaruroro
Te ika i Tukituki, te ika i Porangahau
Te ika i Te Whakaki te takina mai ki te turuturu
Ki tenei tapangutu, ki tenei tauremu
I whiwhia mai a Tangaroa, Tangaroa whiwhi
E tuku, e heke ki to moenga
Ki tenei kupenga, ki tenei tauremu
Ana oti kai a koe; whiwhia, rawea.

Fish-weirs, especially eel-weirs, were highly valued in pre-European days, and any interference with such by unauthorized persons would at once cause trouble, a furious scene of noisy quarrelling, in which lives might be lost.

Some description of the Whanganui form of weir, the pa auroa, should here be given, though such account cannot equal that compiled by Mr. Downes which was published in vol. 50 of the Transactions of the New Zealand Institute. Perhaps Captain Mair was the first to give us some description of this form of weir, and it appeared in the Transactions of the New Zealand Institute for 1879. It runs as follows: "During the summer and autumn rains large quantities of fish are caught by the natives in weirs, or single lines of stakes driven into the bed of the river at the heads of the rapids, and placed at an angle of about 10 degrees across the current. These lines of stakes are 50 ft. or 60 ft. long, and 15 yards or 20 yards apart, according to the width of the river. Hinaki, or wicker baskets, are placed at the bottom of each row of stakes, and the fish, which appear always to swim down the middle of the river, upon meeting these lines of stakes placed at a slight angle to the stream, follow them down into the funnel-mouthed hinaki, and are thus caught. As soon as the pot is full it is replaced by an empty one. In this way I saw about 7 cwt. obtained from two hinaki at Whenua-tere, on 25th February, 1879. The fish so obtained are papanoko, toitoi, inanga, atutahi, upokororo, and a peculiar kind of eel called tuna-heke." Of the blue-eyed species of eel called tuna-heke Captain Mair writes: "The natives keep them in large wickerwork baskets, placed in the small streams, for many months, feeding them upon boiled potatoes, which keep them in good condition." (See figs. 39, 40, 41, 44-48.)

The illustrations given will serve to explain Captain Mair's remarks. In the case of a three-fence weir photographed at Rangitoto, Hiruharama (Jerusalem), on the Whanganui River, the three fences were certainly much closer together than the 15 yards mentioned by the page 146above writer, and I cannot help thinking that "yards" in his narrative should read "feet." (This at the down-stream end of the fences.) Natives like to erect these weirs in the middle of the river because the eel-run is principally confined to that part, and to put three fences 15 to 20 yards apart would call for a wider river than Whanganui in order to preserve the desired arrangement.

It will be seen that these three fence-like barriers are really three separate weirs, there being no connection between the three fences. The stream runs free between them, as also between them and the two banks of the river, thus providing wide loopholes of escape for Tangaroa ara rau. In the case of a V-shaped weir the wings of the V are continued to the banks of the stream, and the whole of the waters of the stream flow through the ngutu outlet (or outlets) and through the wattled fences.

The three weir-fences at Rangitoto were set at a very acute angle to the current, and at slightly different angles to each other, occupying the middle part of a rapid between two calm reaches. I gathered from the natives that these different angles are deemed suitable to the different phases of a freshet. Other such weirs have but two of these fences.

page 147

The illustrations show the form of these weirs and the mode of construction. The name of the upper horizontal ricker of a weir-fence was given me as raumatangi, while ngakau and uaua (huahua) denote the lower ones. The struts bracing the fence are noko, and the offset post at the down-stream end the pou-tihoi. The photographs will serve to show what heavy horizontal timbers are employed in order to impart stability to these weir-fences. The forked-branch section that strengthens the hoop of the funnel-net is termed the teka, and the two horizontal pieces against which it rests are rango. (See fig. 39, p. 137.)

In coming down-stream the eels coming in contact with the upper end of the fence will follow it down-stream and so pass into the net that conducts them into the eel-pot. Mr. Downes gives the length of the fences as being 50 ft. to 60 ft. The offset posts are set about 5 ft. from the fence, and are braced as shown in fig. 40. The rango, or horizontal cross-pieces, take the strain of the net and attached eel-pot, for the teka, which is lashed to the hoop of the funnel-net, and braces it, is so placed as to rest against the up-stream side of the two rango horizontals. (See fig. 39, p. 137.) This can be lifted at any time when the operators wish to empty the pot. All lashing-material consists of tough, pliant stems of climbing-plants. Downes speaks of two pots being set, one above the other, during a freshet, and of the rango (?the lower one) being lifted together with the teka and hoop by means of cords, as shown in his sketch.

We now come to the most important matter connected with our eel-weir, and that is the depositing of the mauri, or talismanic symbol that brings the weir under the care and protection of the gods. This mauri is a protective symbol, and served, in Maori belief, as a form of abiding-place for the atua (spirit gods, supernormal beings) whose aid was solicited to render the weir a successful one. The mauri was in most cases a stone, probably because stones are durable, and it was always concealed, lest persons of evil designs should find it and deprive it of its powers. It was often placed underwater at the base of one of the posts at the ngutu, or outlet of a weir—the first post set up there, which was the right-hand one. All such talismans were treated with respect, and, at least in the case of the mauri of a forest, certain simple ceremonies to ensure good luck were performed at them.

The natives of the Whanganui valley term this talisman an iho, a word that means kernel, innermost part, &c., and they informed me that they had been in the habit of concealing it away from the weir. A stone and a piece of the lashing-material used in the construction of the weir were taken to serve as a mauri and concealed page 148 page 149 page 150 at some place away from the weir. It was advisable, I was told, to bury such an object near a waterfall or cascade, so that it could not hear any destructive magic spells uttered by malicious persons. This explanation seems to show a belief that the spells designed to effect the beneficial powers of the talisman are not potent unless they are heard by the objective—surely a quaint notion. One often finds oneself in contact with superstitious beliefs when among natives, even in these days. When a European wished to photograph some natives engaged in setting eel-pots at a weir he was not permitted to do so, for they feared that the luck of the weir would be seriously affected.

In his work, Tutira, Mr. H. Guthrie-Smith makes some remarks on eels. He speaks of there having been sixteen eel-weirs in half a mile of a creek at Tutira. He gives hinaki ruru as a term for (apparently) a corf in which to keep captured eels alive. He also gives a description of a whare tuna, a kind of eel-refuge, a place in which they lived under desirable conditions and so could be found when required for the serious purposes of life. The writer gives the impression that this was at least a local usage practised at various places; it has not, however, so far as I am aware, been reported from other districts. This accommodation-house for eels was constructed in a creek, and was, say, 15 ft. long, 4 ft. wide, and 1½ ft. high. The sides, roof, and ends were made of manuka (probably brush), lashed with flax. The stream flowed through this enclosure, and within it was placed a quantity of vegetation (water-seeds) to serve as cover for the eels, which are said to have taken possession of it in numbers. Presumably they used it as a kind of headquarters, or permanent home to which they returned after food-seeking forays. When it was desired to take eels from this preserve the hoop-distended mouth of a purangi net, already described, was placed at the entrance to the enclosure, and to the small end of such net an eel-pot was attached as usual. The eels in the enclosure were then induced by divers means to endeavour to escape from the enclosure, when they passed down the net into the eel-pot. The local natives informed Mr. Smith that they had been led to form these eel-shelters by observing that eels are often found in hollow logs in the beds of streams. The eels taken at Tutira are the riko, tatarakau, and pakarara.

Some explanation of the use of a net at an eel-weir, in place of the eel-pot, has been given above, as contributed by Hakaraia Pahewa, of Te Kaha. The funnel-shaped net used for this purpose was, in its lower and smaller end, made with small meshes to prevent the escape of the eels. When these once entered the net the swift rush of waters kept them at the lower (or down-stream) end thereof until they page 151were taken out by the weir-tenders. The force of the current prevented them swimming against it. This mode of taking eels is known as whakaheke tuna.

The purangi, or funnel-shaped net used with an eel-pot, was made of strips of undressed Phormium leaf, but I have been told that occasionally, a net for this purpose was made of dressed and twisted Phormium fibre, being known as a tawiri. As differing names and usages are often encountered in different districts, confusion often arises, and I still believe that such nets of undressed material are termed tawiri in some places. Ngati-Porou state that this guiding-net was used in the deeper streams, but in shallow or narrow streams no such net was used with the eel-pot. The use of the net only at a weir demanded a swift, strong flow of water, otherwise the eels would escape. Mr Downes gives poha as the name of this net at Whanganui, and in some places it is known as rohe. (See figs. 42, 43, 43A.)

Differences are noted in methods of taking eels in different districts, and also in the quantities taken. Extensive operations were formerly undertaken in districts where eels were very numerous, and taken either in rivers, as at Whanganui, or in lagoons and swamps, and their outlets. A considerable amount of labour was necessary in some cases, as in constructing the heavy weirs of the Whanganui River in the days of rude stone tools. At some places but little labour was expended in weir-building, as when eels were taken in small streams or narrow lagoon-outlets. In these streams, of a width of but a few feet or yards, but little weir-building was necessary. This is noted in a brief contribution from Tamati Ranapiri, of Ohau, for which see Appendix 5. This correspondent makes an interesting remark anent the curious custom of allowing eels to run free during the first few nights of their migration to the ocean. He gives three nights as the free-run period in his district, after which they would be trapped. The belief was that if eels were taken during the first nights of the run, then the migration would cease abruptly, although the freshet water was still coming down. Now, was this purely a superstition, or was it the result of caution on the part of the directing experts? Perchance they feared to take too great a proportion of the migrating eels lest a poor run result later on. They knew that eels go to the sea to breed.

Our correspondent also refers to the fact that, of the first lot of eels taken during the season's run, two separate lots were cooked—one for the men, and the other for the women and children. This would be a ceremonial usage.

The following charm, we are told, was repeated when the whatu was placed on or above the net, but no explanation is to hand as to page 152what the whatu is; it may be the mauri of the weir:—

Te whatu kai te mana, te whatu kai a koe na
Ma te paihau mai to ara titi korure
Ma te whakareinga mai to ara titi korure
Ma te paepae mai to ara titi korure
Tangaroa whiwhia, Tangaroa rawea.

This was given by Ngati-Awa of Te Teko.

A South Island native contributed the following account of how eels were taken on the east coast of that Island. These data appear in the original Maori in Appendices 13 and 14. The word whaitau, employed in the description of the drying process, is new to me, and possibly denotes the practice of arranging eels side by side when suspended on the drying-racks. The kaitara device for taking eels is also unknown to me—at least, under that name; it appears to be a form of trap. In his account of the taking of eels by means of eel-pots set at weirs our South Island contributor gives a crude sketch showing two forms of the V-shaped weir. One of these is the ordinary form, in which the apex of the weir is down-stream and represents the outlet at which the eel-pot is set. In the other form the apex is upstream and is closed. An aperture thereat would be almost useless for taking eels moving down-stream. In this case two apertures are left for traps—one at either side of the stream—at the down-stream ends of the diverging weir-fences. In the event of there being shoal water near the bank, there would presumably be a short wing fence running out from the bank in order to provide a sufficient depth of water in which to place the eel-pot. Here follows a rendering of these Ngai-Tahu notes:—

An Account of the Periods during which Eels were taken, and of the Methods employed.

Let the first method be explained; it is one that was used at all seasons, and was known as the rauiri method. A weir was constructed in a stream; eel-pots were set at it in the evening, and in the morning they were lifted, full of eels, brought to land, and the eels in them were killed. Should large eel-pots be used, then as many as a thousand eels, more or less, might be taken in one night. The eels were hung up on a rack and partially dried; they were then taken down, cut open, strung, and again suspended from the racks, where they were frequently turned; when perfectly dry the turning was discontinued and they were just allowed to hang. They were not taken down, but allowed to hang in masses until wanted for a feast or as a present for certain chiefs of other villages.

page 153

Eels taken by hand: Another method by means of which the Maori took eels was by a person, or persons of different clans, seeking them in the streams. In daytime all eels are within their holes. A hand thrust into the hole would feel the eel. Should the eel be a large one, a horepara or kouka, then the fingers of the hand were thrust into the gills of the eel, and so it was brought to land. There might be as many as twenty eels, more or less, in one hole. Persons of all communities followed this procedure, but people of one clan would not go eel-hunting on lands belonging to another clan; on no account would they do so. This method of taking eels was one always followed, and eel-frequented streams were handed down from one generation to another.

Spearing eels by torchlight: Another method of taking eels was seeking them at night by torchlight and spearing them with a matarau. One person might thus take two or three strings of eels in a single trip. When the torches were burned out, then fishers would return. Always in taking eels the Maori folk were careful not to trespass on lands of other clans.

Eels taken in nets: Another way of taking eels was by netting them. Each clan had its own netting-grounds. Waihora was the lagoon where nets were used all round its shores, and such netting-places were passed on from one generation to another.

The koumu method of taking eels: Yet another method of taking eels was as follows: In the autumn eels proceed to the outlet of the lake-waters, and there people go to secure them. The plan was to excavate channels. The place where these were dug is not earth, but gravel. These channels were perhaps 3 chains long and about 4 ft. wide, and the fresh waters of Waihora were allowed to flow through them. The water was constantly seeping through the gravel and flowed readily down the channels. This method was made use of on dark nights, not when the moon was shining; were the moon shining, then the eels would not enter the channels, but only on moonless nights. Eels would of their own accord enter the channels when water was flowing through them. Men engaged in attending to these devices remained alert throughout the night. On nights when many eels entered the channels the mouths of such channels were blocked with stones and gravel, or with net traps called kohau, in order to prevent eels passing out of the ditches. The water-flow being blocked, the channel became dry, and so the eels therein were easily caught and confined in the kohau^*. Should the eel-run be a good one and page 154continue throughout the night, then two or three thousand eels might be taken in a single channel in one night. This would be on nights when many eels were running; some nights would see the numbers much reduced. During rain they would be very numerous and run freely. Sometimes the surf broke right over the gravel-bank between Waihora Lagoon and the sea; at such times eels collected to lick up the foam of the breakers; at such times multitudes of eels could be seen, and they could be taken in the daytime, all through the night, and, in fact, until the surf ceased to wash over; then only would eels cease to haunt the gravel-bank. The wholesale taking of eels at these times is alluded to by natives as tai whati and tai pupu, an allusion to the breaking waves.

Another way in which eels were taken at Waihora was when the shoal waters of the lagoon were much affected by strong winds. The result was that a considerable area of the lake-bed was exposed and many eels would be seen wriggling on the mud. The lake-bed was so affected for perhaps a mile. When natives heard a strong wind blowing they knew that soon the waters of Waihora would recede; hence they would proceed to the lagoon and wait until the bed was exposed. As the waters receded, the people advanced and found the stranded eels wriggling about on the mud. Then all they had to do was to string them on cords by means of a bodkin-like implement, and so fill cord after cord. But ever they warily watched for the returning waters. Ere the waters came sweeping back the eel-collectors would return, dragging their strings of eels after them in an energetic manner, lest they be caught by the waters and drowned. Some natives have so lost their lives at such times, for should the wind die away quickly, then the waters swiftly returned, and should any persons be overtaken by them they would be submerged. Then it was a case of swimming, and those who were poor swimmers perished, while expert swimmers survived and reached land. Each man would use two stringing-cords at this task; these were perhaps 30 ft. long, more or less; perhaps as many as three hundred eels would be strung on each cord.

Eel-bobbing: Another way of taking eels was by means of bobbing, worms and other bait being used. This method was followed at appropriate times. But this mode of fishing did not result in large numbers being caught, but merely such numbers as could be obtained by such a method, which is followed in streams and lagoons.

Eel-pots: Another way of taking eels was by making a kaitara, in which bait was put, and stone sinkers were secured to that kaitara, when it was lowered into the waters of a stream or lagoon; the cord attached to it had one end secured on the bank. These traps were page 155set in the evening, and lifted and examined in the morning. In some cases the trap would be full of eels; sometimes but few were so taken.

There were other methods for taking eels—indeed, the taking of eels was quite an important industry. All the methods of taking them described above were practised up to the time when the work ceased. When a certain method ceased to be employed, then another was followed for perhaps two or three months, after that yet another method, and so on during the whole year. Still, summer was the principal season for taking eels, and, indeed, for procuring many food products to serve as supplies for feasts and gifts, also as winter supplies for the different villages.

How the eels called hao, or mud-eels, were taken: The task of taking this fish, the hao, was commenced in autumn, in the month of January, that month being, in Maori reckoning, the beginning of autumn, which ended with April. The Maori tale of months commenced in May—

Matahi . .	..	May	Ono (the sixth)	..	October
Maruaroa . .	..	June	Whitu (the seventh)		November
Toru (the third)	..	July	Waru (the eighth)	..	December
Wha (the fourth)	..	August	Iwa (the ninth)	..	January
Rima (the fifth)	..	September	Ngahuru (the tenth)		February

January is included in the term Ngahuru. The month called the Matahi by natives was the commencement of the new year of the Maori.

The hao is found in swampy ground; it abides in large swamps. It is a species of eel, but a small one, and has the form of an eel. When heavy rains came, then persons accustomed to the work knew that the hao would be abroad; hence they proceeded to their eel-weirs and set the eel-pots. In the morning those traps were full; sometimes they were so filled two or three times in one night. The trap-tenders remained at the weirs during the night when the eels were running freely, in order to attend to the traps. Some of the eels so taken would be dried for future use; others would be cooked. In drying eels a large fire was kindled and allowed to burn down to a mass of embers, over which the eels were placed on a grid made of green rods. This rack was carefully constructed of stems of young manuka by the Maori. Those persons who understood such work made those erections for the purpose of drying the eels. Such work was done by skilled persons, even from olden times; our ancestors handed these methods down from one generation to another. When the eels were so dried they were hung up on stages used for such purposes. These tasks began in the tenth month and continued page 156 until the Matahi month, when the taking of these eels ceased, and they were taken to the villages, fortified and open. Here end these South Island notes,

Angas, in his work, New Zealanders Illustrated, gives in Plate 55 a representation of an eel-weir on the Mokau River, here reproduced in fig. 49. It depicts a singular and puzzling erection. Two high palisade fences are seemingly situated at the sides of the stream, at the water's edge; these look high enough for a deer-drive. Across the stream a net seems to have been secured to a horizontal rail in a permanent manner. The observer fails to see how an eel-pot could be set at this remarkable weir to be of any service. We have no other information concerning this weird weir, and so must set it aside as "not proven." It seems to be a kind of frightful example of what an eel-weir should not be.

Reverting to the remark concerning the dragging of a catch of eels, this was a common usage when natives were taking eels in shoal lagoons, &c.: the eels were strung together and dragged by the eeler. When busy with spear or hand catching, the operator would sometimes secure the cord to one of his legs, and so pull his catch after him. A string of a small number of small eels might be carried suspended from the arm. (I Poukawa a Ngai-Tangihia e noho ana, he kaikape i nga tuna te take i panaia mai e Te Manawa-kakawa i Poukawa. Ko page 157nga tuna nunui ka tui, ka here ki nga waewae to haere ai; ko nga mea pakupaku ka tui ki nga ringa mau haere ai. Koia nei te take i panaia mai i reira, ka heke mai nei ki te Waikopiro nei noho ai.)