Concerning The Collembola Tullbergiinae
Publication of this paper is assisted by a grant from the Victoria University of Wellington Publications Fund.
A review is given of the species of Tullbergiinae described by R. S. Bagnall in a series of papers published in 1935-37 and 1947. A new species of the genus Tullbergia is described from Australia.
This paper is the sequel to my paper of 1959 which dealt with the Onychiuridae of the late Dr R. S. Bagnall's collections of microscope slides of Collembola in the British Museum (Natural History), London.
Examination of Bagnall's type specimens of the Tullbergiinae has led me to extensive revision of some of his species. This is set out in the following pages and figures drawn from his material, and completes the revision commenced in my 1959 paper. The same criticisms apply to Bagnall's work on the Tullbergiinae as I recorded in the introduction to my 1959 paper.
I would like to record once again my thanks to Dr T. Clay and the Director of the British Museum, National History) for their assistance with material and the many concessions extended to me over the exceptionally long period of this study.
The invalidity of the genus Protullbergia Bagnall, 1947
When Bagnall erected the genus Protullbergia, he chose Tullbergia trisetosa Sehaeffer as the type species, basing his separation on such variable characters as the pseudocelli, unguiculus and antennal base. These characters do not warrant generic distinction, and hence Bagnall's genus must be rejected as a subjective synonym of Tullbergia Lubbock, 1876.
Of the three new species described by Bagnall as belonging to his Protullbergia only one, P. womersleyi, is valid. Examination of Bagnall's specimens show that P. salmoni Bagnall is identical with Tullbergia australica Womersley, 1933, and P. willemi Bagnall is identical with Dinaphorura diversispina Womersley, 1935.
Tullbergia womersleyi (Bagnall, 1947)
Protullbergia womersleyi Bagnall, 1947
I have examined Bagnall's material and this is quite clearly referable to Tullbergia. The SO Ant III is as shown in Fig. 3 with three "laminated" superior sense rods only. Ant IV bears a small apical sensory knob, 6-8 moderately long curved sense rods and numerous long setae. PAO has 60-80 vesicles of the shape shown in Fig. 2 arranged as two parallel rows. The anal spines are 3-4 times as long as broad at their bases, on papillae, and the spines subequal to the hind claw in length. Pseudocelli are not as given by Bagnall but are as follows: 1 + 1 on each of antennal base, hind margin of head Th I — III and Abd I — V. Claw as shown in Fig. 1 with a basal seta and well developed unguiculus. Clothing moderate of short simple setae, longer posteriorly.
Tullbergia australica Womersley, 1933
Protullbergia salmoni Bagnall, 1947
Figures 5-9 are drawn from Bagnall's paratypes of P. salmoni and show that the characteristic body structures of Bagnall's species are so similar to those of T. australica Womersley that it is impossible to seperate these two species. The PAO of Bagnall's specimens have 55-63 tubercles and the SO Ant III agree exactly with that of Womersley's species with the third sense rod on the opposite side from the pair, and fully exposed. Womersley's name published in 1933 must therefore take priority.page 3
Figs. 1-4 Tullbergia womersleyi Bagnall, 1947. Drawn from Bagnall's paratypes.
Fig. 1 hind claw
Fig. 2 end of PAO
Fig. 3 sense organ Ant III
Fig. 4 anal spine and papilla
Figs. 5-9 Tullbergia australica Womersley, 1933. Drawn from paratypes of Protullbergia salmoni Bagnall, 1947.
Fig. 5 two portions of PAO
Fig. 6 claw from side
Fig. 7 main portion of sense organ Ant III, the third sense rod not shown
Fig. 8 apex Ant IV and sense organ Ant III
Fig. 9 anal spine and seta
Scale: A for all Figs.
Tullbergia schaefferi n.sp.
Amongst Dr Bagnall's slides were two of a species labelled T. schaefferi Bagn., the description of which I cannot trace in the literature. These slides are labelled "Potonga Bay, (L. Harrison), VI, 14, Australia" and "Lindfield (L. Harrison) VII, 14, Australia" respectively, and apparently the species has never been described. Accordingly I am describing it now under Bagnall's original manuscript name. It is most similar to T. australica Womersley differing from that species in the long whiplike terminal bristle to the unguiculus of the fore and middle feet and in the sense organs of Ants III and IV.
Both specimens had been badly mounted and squashed but nevertheless all taxonomic details were still remarkably clear enabling the preparation of a reasonably full description and drawing.
Colour: apparently white.
Clothing: sparse of short simple setae longer posteriorly.
Body: length 0.5 mm. Antennae with the segments indistinctly separated; Ant IV with apical knob, and five large stout bent sense rods. There are numerous long setae associated with these sense rods while the setae on the opposite surface are short. The three sense rods of SO Ant III are fully exposed, approximately in line, with a pair of short sense clubs protected by very low chitinous ridges lying between two of the sense rods; the whole with five guard setae (Fig. 12). There is a transverse band of long setae around Ant III just below the apex. PAO with 64-68 tubercles arranged in two parallel rows, but not necessarily with the tubercles exactly opposite one another; each tubercle with a distinct basal socket; the distal foliaceous lobes irregular in outline and sometimes overlapping (Fig. 14). Pseudocelli present but impossible to tabulate on these damaged specimens. Mandible normal with large molar area and head bearing four teeth, three being large and one small. Maxilla head with large base bearing two lobes, one of which carries two large curved teeth and a smaller projection, the other tooth large and curved. There are two serrated shafts as shown in Fig. 13.
Anal spines stout, curved, on short papillae; the spines about half as long as the claws of the hind feet, and guarded by three anterior and three posterior setae each in transverse rows (Fig. 10).
Legs: claws as shown in Fig. 11, without any teeth, basally finely granulate: unguiculus bulbous and finely granulate basally, with a narrow outer lamella and the shaft projected as a long whiplike bristle overreaching the claw tip on the front feet, about reaching the tip on the middle feet and somewhat shorter on the hind feet. A moderately long basal seta to each side of claw.
Holotype (Potonga Bay specimen) and paratype (Lindfield specimen) mounted on slides in Bagnall collection, in British Museum (Natural History), London.page 5
Paratullbergia concolor Womersley, 1930
Paratullbergia carpenteri Bagnall, 1935
Paratullbergia womersleyi Bagnall, 1935
I have examined the type material of P. concolor Womersley and of Bagnall's two species P. carpenteri and P. womersleyi. I am satisfied that all three are the same species for which the name P. concolor Womersly must take priority.
The main variation of this species lies in the number of vesicles in the PAO which can range from as low as 20 to as high as 80. This variation can be traced through the material which was available to me both from the British Museum and from my own collecting in Britain.
P. concolor is generally white, to creamy-white or greyish white in colour. It varies in length from 0.8 mm to 1.2 mm. The clothing is moderate of simple setae generally longer around Abd VI, with a tendency for a transverse row of short setae to occur around the posterior margins and a similar band of slightly longer setae around the anterior margins of each segment. Ant IV (Fig. 19) with up to four curved sense rods, a very small superficial sense rod in a pit, and clothed with short simple setae.
The sense organ of Ant III consists of a pair of very large rectangular to kidney-shaped sense clubs on short rod-like bases, situated behind a low tuberculate cuticular fold with a short central guard seta, three longer lateral guard setae and a single short straight exposed sense club to one side (Figs. 20, 22, 27). The PAO with 18-80 contiguous vesicles primarily lying at right angles to the long axis of the organ, and appearing superficially as in Fig. 18. Under closer examination the PAO shows prominent basal lobes of attachment as in Fig. 16.
Mandible with three strong apical teeth and well developed relatively small molar area (Fig. 26). Maxilla head with three apical teeth and three toothed lamellae (Fig. 23). Pseudocelli very small and obscure appearing as in Fig. 17. Abd VI and the margins of the other abdominal segments with a tendency to coarser tuberculation of the cuticle than the remainder of the body. Abd VI bears two large, curved anal spines on papillae, touching at their bases. Two chitinous tuberculate ridges occur on each side of the anterior margin of Abd VI; these are probably what Bagnall called "tooth-like projections near base". The posterior of these each carries a long simple seta on the outer edge (Figs. 25, 28).
Foot with simple untoothed claw having a fairly long basal seta to each side. Unguiculus vestigial and spine-like (Figs. 21, 24).page 7
Figs. 16-28 Paratullbergia concolor Womersley, 1930.
Fig. 16 PAO surface appearance, P. carpenteri type spec.
Fig. 17 pseudocellus, P. carpenteri paratype spec.
Fig. 18 PAO deep appearance, P. carpenteri paratype spec.
Fig. 19 apex of Ant IV, P. carpenteri paratype spec.
Fig. 20 side view of sense organ Ant III, P. concolor type spec.
Fig. 21 hind foot, P. concolor type spec.
Fig. 22 frontal view of SO ANT III, P. concolor type spec.
Fig. 23 maxilla head, P. womersleyi type spec.
Fig. 24 hind foot, P. womersleyi type spec.
Fig. 25 Abd VI dorsal view, P. carpenteri paratype spec.
Fig. 26 mandible, P. womersleyi type spec.
Fig. 27 lateral oblique view SO Ant III, P. womersleyi type spec.
Fig. 28 lateral view anal spine, P. womersleyi type spec.
Scales: A for Figs. 16, 21 and 27; B for Figs. 7, 17, 18, 19, 20, 22, 23 and 28; C for Figs. 24, 25 and 26.
Dinaphorura diversispina Womersley, 1935
Protullbergia willemi Bagnall, 1947
Dinaphorura harrisoni Bagnall, 1947
I have examined Bagnall's paratypes of P. willemi and his type and paratype of D. harrisoni and am satisfied that both of these species are conspecific with D. diversispina Womersley, 1935.
Austraphorura wahlgreni Bagnall, 1947
The type specimen in the British Museum is badly mounted and crushed as mentioned by Bagnall in his original description. In my opinion a new genus and species should never have been accepted into the literature let alone described from a single crushed specimen. Nevertheless, sufficient of the characters are visible to validate it and to permit the preparation of the accompanying figures. The genus is characterised by the large, stout, rather blade-like, transversely directed sense rod of SO Ant III. This overlies a chitinous fold with a pair of minute sense clubs. A further sense rod lies somewhat removed and below the whole SO with five long guard setae as in Fig. 34. The PAO has 18 detached vesicles as shown in Fig. 36.
Pseudocelli with 1 + 1 on antennal base and Abds IV and V. There may be more on other segments but because of the condition of the specimen they cannot be determined. The apex of Ant IV may have a small retractable sensory knob. Sensory setae of Ant IV could not be determined.
The foot with a stumpy vestigial unguiculus and basal seta, no teeth (Fig. 35). The anal spines long, slightly curved, on papillae as in Fig. 37; the papillae about one third length of spines and the spines themselves subequal in length to hind claw.
Clothing of moderately long simple setae, longer posteriorly.
Neotullbergia laingi Bagnall, 1936
Two specimens, a type and a paratype, were in Bagnall's collection. The type is considerably damaged and only sufficient is visible to determine the validity of the genus. The paratype, although also a poor specimen, was used to draw the figures presented here.
Ants III and IV appear to be fused with the antennal segments related as 8: 10: 24. Ant IV has a small apical sensory knob, four-five sensory rods and numerous simple setae. SO Ant III with four sense rods, two sense clubs behind a cuticular fold and eight guard setae, all disposed as in Figs. 39 and 41. The sense rod shown in Fig. 41 is situated on the opposite side of the segment from the main part of the sensory organ. PAO of about 40 irregular shaped vesicles arranged more or less in four rows in a finely granulate area, clearly demarcated from the surrounding coarser cuticular granules. Pseudocelli impossible to decipher.
The clothing is of sparse simple setae, longer posteriorly. Length 1.0 mm., colour apparently white.
Abd IV dorsally bears two large branched anal spines, langer than the hind claw, on tuberculate papillae; anterior to these are two tuberculated chitinous ridges, the whole with setae disposed as in Fig. 42. The anal spines are four times as long as their papillae and their basal portions are also tuberculated. The branches arise basally, one dorsolateral and one ventral. These structures characterise the genus.page 10
Figs. 29-33 Dinaphorura diversispina Womersley, 1935. Drawn from paratype specimens of Protullbergia willemi Bagnall.
Fig. 29 Abd VI
Fig. 30 PAO
Fig. 31 apex Ant IV
Fig. 32 sense organ Ant III
Fig. 33 hind foot
Figs. 34-37 Austraphorura wahlgreni Bagnall, 1947. Drawn from type specimen.
Fig. 34 sense organ Ant III
Fig. 35 hind foot
Fig. 36 PAO
Fig. 37 anal spines
Figs. 38-42 Neotullbergia laingi Bagnall, 1936. Drawn from paratype specimen.
Fig. 38 PAO
Fig. 39 sense organ Ant III
Fig. 40 hind foot
Fig. 41 fourth sensory rod Ant III
Fig. 42 anal spines
Scale A for Figs. 30-32, 36, 38-39, 41.
Scale B for Figs. 29, 33-35, 37, 40, 42.
Metaphorura bipartita Handschin, 1921
Metaphorura boerneri Bagnall, 1936
The type and several paratypes of Bagnall's M. boerneri together with M. W. Davies' specimens identified by Bagnall as M. bipartita Handschin and some specimens labelled by Bagnall as M. denisii, but never published as such, were examined. All of these, I am satisfied, are the same species and must be known as M. bipartita Handschin. The accompanying illustrations are drawn from these specimens as set out in the legend to the figures. Bagnall separates his boerneri from bipartita primarily on the presence of pseudocelli on Th I but, though Handschin did not observe these, they have been recorded by others, notably Stach (1954, p.211). This description by Stach is very full but I would add the further notes that in Bagnall's material Ant IV has six-eight stout curved sense rods and a small apical knob (Fig. 43). The SO Ant III sits behind a high cone-like integumentary fold with a papilla at each end (Fig. 45 and 51). Behind the cone-like fold is a further low fold in which are two small sense clubs flanked on each side by a large, inwardly curved, stout sense rod (Figs. 57 and 49).
There is always a third papilla to one side (Figs. 45 and 51) and a third, stout, curved sense rod on the opposite surface of the antennal segment (Fig. 43). The SO proper has four guard setae (Fig. 57). PAO with 20-25 "U" shaped vesicles (Figs. 46 and 55) situated in a deep furrow and hence difficult to see. The vesicles appear to change in shape as focus of the microscope reaches deeper into this groove (Fig. 54).
The papillae of the annal spines are not always distinct but are generally recognisable by the smaller size of the cuticular granules as compared with those dorsally on Abd VI.
Figs. 43-57 Metaphorura bipartita Handschin, 1921.
Fig. 43 Ants III and IV, M. boerneri type spec.
Fig. 44 foot, M. boerneri type spec.
Fig. 45 SO Ant III cone-like fold removed, M. boerneri paratype spec.
Fig. 46 PAO M. bitpartita Bag. M. W. Davies' spec.
Fig. 47 Abd VI showing anal spines and large ventral tubercle, M. denisi spec.
Fig. 48 genital aperture, M. boerneri type spec.
Fig. 49 sense clubs and rear fold of SO of Ant III, B. boerneri paratype spec.
Fig. 50 anal spines and papilla, M. boerneri type spec.
Fig. 51 SO Ant III surface appearance, M. boerneri paratype spec.
Fig. 52 Abd VI anal spines and papilla, M. boerneri type spec.
Fig. 53 genital aperture, M. boerneri paratype spec.
Fig. 54 end of PAO deep view, M. boerneri paratype spec.
Fig. 55 end of PAO surface view, M. boerneri paratype spec.
Fig. 56 genital aperture, M. boerneri paratype spec.
Fig. 57 SO Ant III, M. boerneri paratype spec.
Scales: A for Figs. 43, 44 and 47; B for Figs. 49, 51, 54, 55 and 57; C for Fig. 50; D for Figs. 52 and 56; E for Figs. 45, 46, 48 and 53.
Neonaphorura duboscqui Denis, 1932
Tullbergia duboscqui Denis, 1932
Neonaphorura duboscqui Bagnall, 1935
Neonaphorura anglicanus Bagnall, 1936
Neonaphorura duboscqui Bagnall, 1936
There are seven slides of this species in the Bagnall collection, six being labelled N. duboscqui and one N. devoniensis. This latter name was never published by Bagnall but in 1936 he did publish the name N. anglicanus for the British specimens he previously had identified as N. duboscqui in 1935. Bagnall's two specimens from Sewerby, E. Yorks, were the ones upon which he based the species N. anglicanus, and five specimens from Petit Tor Bay, S. Devon, were identified as N. duboscqui of Denis. I have examined all these specimens and find them identical in all respects. The specimens from Sewerby are smaller, probably not mature, and cannot be regarded on these grounds as a new species. The postantennal organ of these specimens agrees in every detail with the Fig. 2 given by Denis for this structure in his original description. Denis mentions the small central sense clubs in the SO Ant III as undoubtedly present but difficult to see clearly. I have been able to decipher these structures after remounting the specimens and Fig. 58 shows clearly the relationships of the various parts of this sensory organ.
Mesaphorura krausbaueri Boerner, 1901
Paratullbergia macdougalli Bagnall, 1936
Mesaphorura thalassophila Bagnall, 1937
Mesaphorura baconae Bagnall, 1947
Bagnall's collection contains two slides labelled P. macdougalli, the one labelled "type" and the other "paratype". The type is illustrated in Figs. 65-67 and identifies as M. krausbaueri Boerner. The paratype slide identifies as Protaphorura armata Tullberg. The recording of P. macdougalli Bagnall as a subjective synonym of M. krausbaueri Boerner was first made in my "Index to the Collembola", 1964, p.150, but was not explained on that occasion.
Two slides labelled Mesaphorura, thalassophila in Bagnall's collection were not labelled as type material though they would appear to be those from which his descriptions of this species were made. They identified as M. krausbaueri Boerner and I now record Bagnall's species M. thalassophila as a subjective synonym of M. krausbaueri Boerner.
Similarly a slide labelled as type of Mesaphorura baconae in Bagnall's collection identifies as M. krausbaueri Boerner; this subjective synonymy was first recorded in 1964 on p.151 of my "Index to the Collembola". Fig. 70 is drawn from a specimen identified by Bagnall as M. krausbaueri and which compares in all respects with his P. macdougalli material. Bagnall in his description of P. macdougalli says that "Ant IV is much as in carpenteri" but this is not so as his P. carpenteri I have already shown is synonymous with P. concolor Womersley (Fig. 19).page 15
Figs. 58-64 Neonaphorura duboscqui Denis, 1932
Fig. 58 SO Ant III, Bagnall's spec. N. duboscqui.
Fig. 59 foot, Bagnall's spec. N. duboscqui.
Fig. 60 Abd VI anal spines, Bagnall's spec. N. duboscqui.
Fig. 61 fourth sense rod Ant III, Bagnall's spec. N. duboscqui.
Fig. 62 apex Ant IV, Bagnall's spec. N. duboscqui.
Fig. 63 PAO deeper structure, Bagnall's spec. N. duboscqui.
Fig. 64 PAO surface view, Bagnall's spec. N. duboscqui.
Figs. 65-70 Mesaphorura krausbaueri Boerner, 1901.
Fig. 65 PAO, P. macdougalli type spec.
Fig. 66 sense organ Ant III, P. macdougalli type spec.
Fig. 67 hind foot, P. macdougalli type spec.
Fig. 68 distant sensory rod Ant III, M. thalassophila Bagnall's type spec.
Fig. 69 Abd VI, M. thalassophila Bagnall's type spec.
Fig. 70 Ant III and IV from spec. M. krausbaueri for comparison with type P. macdougalli Bagnall.
Scales: A for Figs. 58, 61, 63, 64, 65, 66 and 68; B for Figs. 59, 60 and 62; C for Fig. 67; D for Fig. 69; E for Fig. 70.
Stenaphorura quadrispina Boerner, 1901
Stenaphorura axelsoni Bagnall, 1935
Stenaphorura denisi Bagnall, 1935
Stenaphorura lubbocki Bagnall, 1935
Stenaphorura absoloni Bagnall, 1936
Bagnall's collection contains: a slide of a poor specimen labelled type of Tullbergia lubbocki which is what he described in 1935 as Stenaphorura lubbocki; slides of a type specimen and three paratype specimens of Stenaphorura denisi; and two specimens on one slide labelled type, and three other specimens on slides labelled paratypes of Stenaphorura absoloni. I could not locate Bagnall's specimen's of S. axelsoni. He based S. axelsoni on Axelson's description (1912, p.96, pl. VIII, Figs. 12-13) of Tullbergia quadrispina and I am satisfied that it is a synonym of S. quadrispina Boerner which was Axelson's original identification.
The illustrations included here are drawn from several of Bagnall's type specimens. Boerner's original descriptions of the PAO and SO Ant III were rather inadequate. The PAO has from 30 to 60, possibly 80, vesicles as shown in Figs. 73 and 78. The organ is situated in a deep cuticular fold and the vesicles are closely packed together. The SO Ant III consists of three bent sense rods and two sense clubs with a cuticular fold and sometimes a papilla as shown in Fig. 72. There is a fourth bent sense rod somewhat removed, usually with four guard setae as shown in Fig. 71 and again in Fig. 76.
Abd VI has cuticular granules noticeably larger than on the rest of the body and bears two large terminal anal spines and two small sub-terminal anal spines (Fig. 74). Th III also has larger cuticular granules dorsally and the claws are finely granulated.
Figs. 71-78 Stenaphorura quadrispina Boerner, 1901
Fig. 71 sense rod Ant III, T. lubbocki type spec
Fig. 72 sense organ Ant III, T. lubbocki type spec
Fig. 73 PAO, T. lubbocki type spec
Fig. 74 Abd VI, P. denisi type spec
Fig. 75 fore foot, P. denisi type spec
Fig. 76 sense organ Ant III, P. denisi type spec
Fig. 77 apex Ant IV, P. denisi paratype spec
Fig. 78 PAO, P. denisi type spec.
Scales: A for Figs. 71, 72, 74 and 75; B for Figs. 73, 76, 77 and 78.
Axelson, W. M., 1912: Die Apterygotenfauna Finlands. II. Spezieller Teil. Acta. Soc. Sci. Fennicae , 40, pp. 1-359.
Bagnall, R. S., 1935: On the classification of the Onychiuridae with particular reference to the genus Tullbergia Lubb. and its allies. Ann. Mag. Nat. Hist. , (10), 15, pp. 236-242.
1935: The British Tullberginae Pt. I. Ent. Month. Mag. , 71, pp. 164-173.
1936: The British Tullberginae Pt. II. Ent. Month. Mag. , 72, pp. 34-40.
1937: Contributions towards a knowledge of the Scottish Onychiuridae (Collembola), II. Scot. Nat. , Edinburgh, pp. 87-90; pp. 145-150.
1947: Contributions towards a knowledge of Tullbergiidae (Collembola-Onychiuroides), I-III. Ann. Mag. Nat. Hist. , (11), 14, pp. 435-444.
Boerner, C., 1901: Neue Collembolenformen und zur Nomenclatur der Collembola Lubbock. Zool. Anz. , 24, pp. 696-712.
1901: Vorläufige Mitteilung uber einige neue Aphorrurinen und zur Systematik der Collembola. Zool. Anz. , 24, pp. 1-15.
Denis, J. R., 1932: Sur la faune francaise des Aptérygotes, XII. Arch. Zool. Exp. Gén. , Paris , 74, pp. 357-383.
Handschin, E., 1921: Die Onychiuren der Schweiz. Ver. Nat. Ges. in Basel , 32, pp. 1-37.
Kos, F., 1940: Terricole Collembola aus Slovenien. Glasnik Skopskog nauenog Razprave , Ljubljana, 3 (11), pp. 137-168.
Salmon, J. T., 1959: Concerning the Collembola Onychiuridae. Trans. Ent. Soc. Lond. 111 (6), pp. 119-156.
1964: An Index to the Collembola. Roy. Soc. N.Z. Bull No. 7, pp. 1-651.
Stach, J., 1954: The Apterygotan Fauna of Poland in relation to the World-Fauna of this group of Insects. Family: Onychiuridae. Polska Awad. Nauk. Inst. Zool. , 1954, pp. 1-219, 27 pl.
Womersley, H., 1930: Some additions to the Collembola of Britain. Ann. Mag. Nat. Hist. , (10), 6, pp. 149-153.
1933: A preliminary Account of the Collembola-Arthopleona of Australia, I: Poduroidea. Trans. Roy. Soc. Sth. Austr. , 57, pp. 48-71.
1935: On some New Records and Species of Australian and New Zealand Collembola. Trans. Roy. Soc. Sth. Austr. , 59, pp. 207-218.
Zoology Department, Victoria University of Wellington, Private Bag, Wellington, New Zealand.