Quarterly Journal of Microscopical Science, Vol s3-88, 275-336, Copyright © 1947 by Company of Biologists

The Development and Affinities of the Pauropoda, based on a Study of Pauropus silvaticus

¹ Zoology Department, University of Melbourne

1. The minute egg is heavily yolked, and is devoid of periplasm or vitelline membrane. Shortly before laying, the germinal vesicle phase gives way to one in which the chromosomes reappear as thirteen ‘tetrads’ in a central clump of cytoplasm. The latter moves to the periphery, and in this first meiotic prophase the egg is laid. The ‘polar bodies’ do not separate from the egg. In most eggs they degenerate very rapidly. Male and female pro-nuclei fuse in the centre of the egg to form a resting nucleus.

2. Cleavage is total and unsynchronized, the cleavage-cells becoming arranged as yolk-pyramids around a central blastocoele. This blastula persists up to about the eighty-cell stage, when a gastrula is formed by migration of one, or at most two cells, from the layer of yolk-pyramids into the blastocoele to form the endoderm.

3. The gastrula is succeeded by a blastoderm of the familiar myriapod-insect type; the cleavage nuclei, except those of the endoderm, move into the peripheral layer of accumulating cytoplasm, while the internal cell-partitions break down. Total cleavage is thereby replaced by a superficial cleavage within the blastoderm. A blastodermic cuticle is secreted.

4. Out of the blastoderm there now differentiates the germ-band; it is of the usual elongate myriapod type, and extends over the anterior and posterior poles of the egg on to its upper half. At no time does a ventral flexure form. The rest of the blastoderm becomes much thinned out and is a provisional body-wall.

5. From the inner surface of the blastoderm isolated cells with enlarging nuclei have meantime migrated as yolk-cells into the yolk; from them will eventually form the fat-body of the larva.

6. During the development of the germ-band the latter becomes the source of the developing mesoderm, cells separating from it in great numbers along its length, and so coming to form a second, and at first very irregular layer, internal to the ectoderm.

7. The early developmental processes are of the usual myriapod type, except for the presence of an easily recognized gastrula. The survival of the latter is of importance for the interpretation of the specialized myriapod-insect type of ontogeny; in particular, the blastoderm phase is found to be a post-gastrula stage, and not a blastula as commonly believed.

8. Stomodaeum and proctodaeum are the first structures to appear in the germ-band. The formation of the head-lobes soon follows. Then the Anlagen of the antennae arise, being at first post-oral in position; and after them appear the Anlagen of the mandibles and maxillae. The premandibular segment does not bear even the rudiments of appendages. The segment behind the maxillary segment is the collum segment, without appendage-Anlagen, and it remains part of the abdomen; there is, therefore, no second maxillary segment. The Anlagen of the first, second, and third legs then appear in succession. Intersegmental lines form only in the advanced embryo; when these eventually appear, they reveal two segments behind the fourth abdominal (third leg-bearing), namely the fifth abdominal and anal segments. The fifth segment of the embryo becomes the fifth of the adult animal, so that the teloblastic formation of new segments must proceed by the budding off of new segments from the stationary anal segment.

9. The embryo now slowly swells, presumably owing to absorption of water, and so outgrows the available space within the egg. A gradually enlarging rent appears in the egg-shell, and after several days the embryo emerges from it in a quiescent ‘pupoid’ phase. Unlike the ‘pupa’ of other myriapods, it shows but little resemblance to the future larva. Within the pupa the first teloblastic segment (sixth abdominal) appears.

10. The development of the abdominal segments, both in the embryo and in the pupa and larva, shows that there are no ‘diplosegments’ in Pauropus; the tergites are derivatives of simple segments, and the apparent ‘diplopody’ arises from the presence of a wedge-shaped segment, with reduced tergal wall, behind the tergite-bearing segments.

11. The differentiation of the head-capsule out of the primitive head-segments is attended by the familiar inturning of the sternal wall of the postoral segments to form the floor of the pre-oral cavity; and by the curving forward of the more lateral parts of the segments towards the front of the head. The antennae are thereby carried into a completely pre-oral position. The pre-mandibular epidermis becomes rolled under to form the roof of the pre-oral cavity (inferior surface of clypeus).

12. The head is composed of a procephalon, in which pre-antennary, antennary, and pre-mandibular segments are represented; and of a gnathocephalon, consisting of only two segments, the mandibular and maxillary. Superlinguae, derived from the mandibular epidermis, are present. The mandibles are closed in by the inturned margins of the clypeus; they are unsegmented and adapted to a diet of semi-fluid food. The maxillae display a cardo, a stipes, and a lacinia. Between them is the intermaxillary plate (sternite of maxillary segment).

13. At no time does the germ-band display a ventral flexure. Embryonic membranes are absent. An embryonic ‘dorsal organ’ appears, but it is not of the type found in Symphyla, Collembola, and Campodea.

14. In the early germ-band the mesoderm tends to aggregate laterally into two bands of cells, from which the succession of somites arises. Although many of the somites soon display very small coelomic cavities, they remain poorly developed, and coelomoducts do not appear. Since there are no blood-vessels in Pauropus, they do not contain any vasoblasts; they are, moreover, quite unique in that they do not even contribute any mesoderm to the mid-gut wall, the splanchnic mesoderm arising entirely from the mesoderm of the stomodaeum. Nor do they participate in the formation of the genital rudiment. The dorso-lateral muscles also are not derived from the somites.

15. Between the rows of somites is a layer of unsegmented ‘median mesoderm’ out of it develops the genital tube, as well as a median band of neuroglia (?) tissue in the nerve-cord.

16. Although the somites are diminutive, a complete set is present. They are the pre-antennary (vestigial), antennary, pre-mandibular, mandibular, maxillary, collum, second, third, and fourth abdominal somites; a fifth abdominal and very small anal somite form in the more advanced embryo. In addition to these there is also a small clump of ‘teloblastic mesoderm’ arising in the late embryo from the mesoderm that is heaped up in front of the proctodaeum; from it is generated the mesoderm of the larva.

17. From the pre-antennary somite arise the buccal dilator muscles; from the antennary and mandibular somites arise the muscles of the antennae and mandibles respectively.

18. From the pre-mandibular somite arises the large pre-mandibular gland; it opens to the side of the mandibles and is evidently a salivary gland. Its duct seems to be of ectodermal origin. There is no associated ‘end-sac’. In Symphyla a pre-mandibular gland is present up to the time the larva leaves the egg, when it degenerates, leaving only its nephrocytes; in Pauropus alone among myriapods is it known to survive, though vestiges of it are found both in chilopods (‘lymphoid tissue’) and insects (sub-oesophageal bodies).

19. From the maxillary somite there develops, in addition to the muscles of the maxilla, the maxillary gland. The latter has an ‘end-sac’, which displays nephrocytic action to trypan blue injected into the blood.

20. The somites of the collum and other abdominal segments, as well as the anal segment, give origin to nothing but myoblasts from which most of the muscles of the respective segments develop.

21. In addition to the glands already referred to, there are present: (a) clypeal glands, that arise from the epidermis of the clypeus; (b) pseudocular glands, lying adjacent to the pseudoculi, and derived from the epithelium of the latter; (c) large intermaxillary glands, derived from the maxillary sternum.

22. The mid-gut epithelium is formed from the endoderm of the gastrula, the cells gradually losing their yolk and slowly increasing in number. The mesoderm of the mid-gut is derived from the mesoderm that is heaped up along the stomodaeum, whence it spreads back as an arching roof to the endoderm, the immediately underlying cells of which become arranged into an epithelium, within which excretory concretions, similar to those of the adult mid-gut, soon appear. The floor of the mid-gut remains for long free from any mesoderm; from it develops a ventral band of enlarged mid-gut cells, permanently free from excretory concretions. The hindermost tip of the mid-gut is of proctodaeal origin. The lumen of the mid-gut does not communicate with those of the stomodaeum and proctodaeum till shortlybefore the larva emerges.

23. The two Malpighian tubes arise from the anterior tip of the proctodaeum; they do not seem to be functional excretory tubes, since in the growing larva they begin to display a markedly degenerate character.

24. The genital rudiment does not arise out of the somites, but from the ‘median mesoderm’ of the fifth abdominal segment, i.e. from the vestige of the mesoderm that remains in front of the proctodaeum after the fifth somites and teloblastic mesoderm have separated from it. Embedded in it is a single primordial germ-cell. A string of cells spreads forward from this mesoderm along the roof of the nerve-cord into the third abdominal segment, thus forming the genital rudiment.

25. The fat-body does not develop out of the somites, as in insects and chilopods, but out of the embryonic yolk-cells, as in Symphyla. It is phagocytic towards injected Indian ink. The haemocoele arises from spaces left by shrinkage and withdrawal of fat-body.

26. The ganglia of the nerve-cord develop in association with ‘ventral organs’, of which there is a single pair in each segment, except the anal segment. In the abdominal segments the ‘ventral organs’ become incorporated into the ganglia. The ‘median mesoderm’ plays an unusual role, in that it gives origin to a median band of neuroglia (?) cells, within the chain of ganglia; there is no incorporation of ‘median cord’ ectoderm into the ganglia.

27. The ventral nerve-cord of the embryo comprises a mandibular and maxillary ganglion (fused into one), followed, in the abdomen, by the ganglia of the collum, second, third, and fourth abdominal (leg-bearing) segments, fifth segment, and a vestigial anal ganglion. The ganglion of the fifth, or penultimate segment, is in reality a teloblastic ganglion, which enlarges, and from which the fifth ganglion proper is separated off in the pupa, the rest remaining as teloblastic ganglion, from which, in the larva, the sixth and remaining ganglia in turn arise.

28. The brain develops-out of: (a) a trilobed protocerebral Anlage, whose posterior lobes invaginate below the surface, but whose lateral and frontal lobes do not invaginate but display ‘ventral organ’ cell-disposition; ((b)) a pair of diminutive pre-antennary (?) ganglia, which give origin to a definite part of the brain between the protocerebrum and deutocerebrum; (c) a pair of antennary ganglia, from which the deutocerebrum will arise; (d) a pair of pre-mandibular ganglia, from which will form the tritocerebrum. The last three all develop in association with ‘ventral organs’.

29. A detailed account of the morphology of the brain and of the cerebral nerves is given.

30. The visceral nervous system consists of: (a) a frontal ganglion, derived from the roof of the oesophagus; (b) a pair of ‘oesophageal ganglia’ (?), which develop in a most unexpected way, for they arise from the inner ends of the mandibular apodemes; (c) a stomachic ganglion, formed from the hinder end of the oesophagus; (d) a ‘caudal’ visceral ganglion, which is apparently the hind end of the last abdominal ganglion.

31. The structure and development of certain epidermal organs (trichobothria, basal antennal sense organs, pseudoculi) is described. The supposed vestigial appendages of the collum segment are not appendages at all, but seem to be related to exsertile vesicles; they develop from vestiges of the ‘ventral organs’ of the collum segment. The hypopharyngeal apophyses are epidermal ingrowths that arise a little in front of, and median to, the mandibles.

32. The greater part of the muscular system is derived from cells that are set free by the break-down of the somites. The dorsal longitudinal muscles are exceptional.

33. Four larval instars precede the adult animal with 9 leg-pairs. These instars have 3, 5, 6, and 8 leg-pairs respectively. The adult animal does not moult.

34. The new segments arise by proliferation of epidermal cells within the anal segment.

35. The mesoderm of the growing zone is generated entirely from the ‘teloblastic mesoderm’, located in the anal segment, successive clumps of mesoderm being allotted to the new segments as these successively develop. The ventral longitudinal muscles always develop precociously. Somites do not form in the mesoderm of the new segments.

36. There is no periodic renovation of mid-gut epithelium in the growing larva, such as is encountered frequently in insects; nevertheless there is considerable cell-proliferation, both in the mid-gut and in the fore- and hind-gut. The Malpighian tubes gradually degenerate, but do not disappear.

37. The fat-body of the larva is supplemented by new fat-body that develops unexpectedly from the epidermis.

38. In the newly emerged larva the genital rudiment, still sexually indeterminate, consists of a narrow string of cells lying between the intestine and the nerve-cord, and extending forward just into the third abdominal segment. Posteriorly it merges with the ‘median mesoderm’ of the growing zone. Its further development in the larva involves (a) slow multiplication of the primordial germ-cells located at its hinder end; (b) posterior elongation of the genital rudiment at the expense of ‘median mesoderm’ cells in the growing zone; (c) thickening of the anterior part of the genital tube. The exit ducts arise as a pair of epidermal ingrowths just behind the second legs, which grow round the nerve-cord and join the anterior tip of the genital rudiment. In the female one only survives, and forms the oviduct and receptaculum seminis; in the male both are retained as ejaculatory ducts. The ovary remains in its primitive position below the intestine, the oocytes arising from laterally placed germ-cells in the median germarium; the anterior end of the original genital rudiment survives in the female as the ‘ductus glandularis’. In the male the gonad-Anlage bends up dorsally to either side of the mid-gut, and divides into four testes; the anterior end of the genital rudiment then splits longitudinally into the four vesiculae seminales. In fourth instar larvae the reproductive organs are often almost mature, but there is no evidence for precocious sexual functioning.

39. The classification of Myriapoda into Progoneata and Opisthogoneata does not seem to reflect the real affinities of the component groups, and in particular it fails badly for the Symphyla which are undoubtedly closely related to the Insecta. In Pauropoda and Symphyla the exit ducts from the gonads are not surviving coelomoducts, as assumed, but epidermal ingrowths, new gonopores having apparently arisen, in adaptation to anamorphosis, remote from the zone of growth. A new classification is proposed, based on the degree of cephalization and specialization of originally abdominal segments: the lowest grade of surviving myriapods are the DIGNATHA (Pauropoda and probably Diplopoda); the Chilopoda, Symphyla, Collembola, and Insecta are TRIGNATHA, and of these the Symphyla, Collembola, and Insecta are united as LABIATA by the common possession of a labium. The Myriapoda seem to have arisen, independently of the other great groups of Arthropoda, from some ancient stock of Peripatus-like ancestors.