Aprils, 1 886] 



NATURE 



547 



wliat is known of tlie anatomy and cmbryolcigy of the various 

 forms. 



In the first place the diagram given on page 123 shows the 

 lines of development of certain groups passing through other 

 existing groups, an arrangement which should be avoided in 

 j^hylogenetic tables. It always suggests that the groups passed 

 through are the direct ancestors of the group at the free end of 

 the line, and it is highly improbable that any existing forms are 

 precisely the same as the ancestors from which a group was 

 derived. To take an example, Uljanin represents the line of 

 evolution of Doiiohnn dcn'-iculatunt passing through first 

 Anchinia rubra^ and then D. viiUUTi^ thus suggesting, I 

 hold, that the two last-named species are extinct forms which 

 were direct ancestors of D. dailktilatuni. Now of course 

 A. jiibra and D. iiiiillerisx^ not extinct, and although they are 

 undoubtedly closely related to ancestral forms on the line of 

 development terminating in D. denticulatutn, still it is very 

 unlikely that they are in all re-pects identical with these an- 

 cestral forms. The best way to represent such a case diagram- 

 matically would be to place A. rubra and D. rmilleri on the 

 ends of short side branches springing from the main stem at the 

 points which the ancestral forms they seem to resemble probably 

 occupied. In that way the line of evolution of Z). denticulatutn 

 would be shown as passing not through, but dose to, A. rubra 

 and D. miilUri. 



Uljanin represents the Appendiculariidse giving rise to the 

 Ascidiae Simplices, from which three lines then diverge, one 

 leading to Sa'fa, the second to the Doliolidae through .4 «<-^z/«'(T, 

 and the third to the Ascidije Compositse. From this last group 

 (presumably only the typical Compound Ascidians) three lines 

 start upwards, leading, — the first to Fjrosomi through Dis- 

 taplia {?), the second to Botryllus, and the third to the " Social " 

 Ascidians through Pieudodidemnum and Tridideinnum. 



The starting-point in this scheme of evolution is undoubt- 

 edly correct. The various existing groups of Tunicata are all 

 descended from common ancestral forms closely resembling the 

 Appendiculariidsc. But the origin of the .Salpid^ and the 

 Doliolida; from groups of the Simple Ascidians is very question- 

 able. The passage from Apficiidiculciria mossi ' through Anrhinia 

 rubra to DoUoluin, and through the ancestral Doliolida; to Salpa, 

 seems so simple and natural that it makes it very unlikely that 

 the Thaliacea were ever fixed .Simple Ascidians which have 

 undergone great modification and have become free-swimming 

 |5elagic organisms again. The argument made use of by Uljanin 

 that the Thaliacea, &c , have not such a typical method of de- 

 velopment as the Simple Ascidians, and are therefore a younger 

 group, is not necessarily of great value, since the process of 

 development may have undergone modification. 



Then, again, it seems more probable that the Simple and the 

 Compound .Ascidians were derived from a common ancestor 

 resembling the simpler forms of the two groups {e.g. Ciona and 

 Diazonn) than that the Compound were derived from the .Simple. 

 IVIany of the Simple Ascidians show far more differentiation and 

 far more specialisation of certain important organs (e.g. the 

 branchial sac in the Molgulidae) than is found in any of the 

 Compound Ascidians. 



I must protest against Uljanin's statement that the " Social " 

 Ascidians are a group derived from the Compound forms and 

 having no close connection with the Simple Ascidians. This 

 view is opposed to all we know as to the veiy close relationship - 

 between the Clavelinida: and the Ascidiidte. There can, I 

 think, be no doubt, after the examination of such a series of 

 forms as Diazona, Clavelina, Ecleiuascidta, and Ciena, that the 

 " Social " Ascidians (Clavelinidas) are intermediate between the 

 least modified forms of the Simple .\scidians {e.g. Ciona) and 

 the least modified forms of the Conpound Ascidians {e.g. 

 Diazona), and ought therefore to be regarded as being closely 

 allied to the ancestral forms from which both Simple and Com- 

 pound Ascidians were derived. 



I agree with Dr. Uljanin in considering Pyrosomx a modified 

 Compound Ascidian, but I differ from him inasmuch as I regard 

 it as being derived from the family Didemnidas, and not from 

 Distaplia. The remarkable new genus Calocormus, found 

 during the CJiallen^er Expedition,"* is a valuable connecting-link 

 between Pyrosoma and the primitive Didemnidre, which in their 



' I have ventured to call by this name a species described as new, but not 

 named, by Dr. Moss (Trans. Linn .'ioc. Land., vol. .\xvii. p. 229, 1871). 



* See Herdman, " Report upon CZ/rt/Zc/i^tf?- Tunicata," part i., and Sluiter, 

 " Ueber einige einfachen .'\scidien," &-., Natuuriiuiidig Tijdsctiri/t v. 

 h'ederlandscli-Indie, Band .tlv. p. i6o, 1885. 



3 See '* Report upon CJtallenger'Xnmza.K:^.," part ii. 



turn were derived from the primitive Distomidas, thus carrying 

 the origin of Pyrosoma back to the typical Compound Ascidians. 



Distaplia, which figures in Uljanin's scheme as the stepping- 

 stone to Pyrosoma, is not really such an extraordinary form as 

 has been supposed. It is an ordinary Compound Ascidian be- 

 longing to the family Distomidae. The Didemnidse {e.g. Tridi- 

 dejnnuni) and the Diplosomidas (t'.^. Pseudodidemnum) are not 

 at all closely related to the " Social " Ascidians. They .ire 

 highly modified and in some respects degenerate Compound 

 Ascidians which have probably originated from the primitive 

 Distomid?e, 



Dr. Uljanin is, I think, right in regarding the Botryllidse as 

 an abnormal group of Compound Ascidians worthy of being 

 placed in a branch by themselves. They show certain resem- 

 blances to some of the Simple Ascidians, and it is just a question 

 whether they should not be regarded as a remarkably modified 

 offshoot from the primitive Cynthiidx quite distinct from the 

 other Compoimd Ascidians. 1 am inclined to regard the Com- 

 pound Ascidians as polyphyletic. There is some evidence in 

 favour of their having arisen as three distinct groups from the 

 ancestral Social and Simple Ascidians, .and in that case one of 

 these groups would be the Botryllida;. This question will be 

 discussed more in detail in a paper I am now preparing on the 

 Phylogeny of the Tunicata 



On some points, then, I am quite in accord with Dr. Uljanin, 

 while we differ on others. With the modifications suggested . 

 above I would accept his views. Probably the most valuable 

 part of his scheme is that dealing with the evolution of the 

 Doliolidae, and the relations between the different forms of 

 DoUolum and Anchinia. W. A. Herdman 



SCIENTIFIC SERIALS 



The Quarterly Journal of Microscopical Science for February 

 l8S6 contains: — On the development of the mole (Talpa 

 iuropiTa) \ the ovarian ovum and segmentation of the ovum, by 

 Walter Ileape, M.A. (plate li). Of the two membranes the 

 outer (zona radiata) was thick, the inner (vitelline) thin ; betvveen 

 them there was a space ; radial canals exist in the outer mem- 

 brane ; no micropyle was observed. Two kinds of yolk are 

 noticed — homogeneous vesicular and minute highly refractive 

 bodies. During maturation the vitellus divided into a medullary 

 and cortical portion, and withdrew from the vitelline membrane, 

 excepting where connected by pseudopodia-like processes, and at 

 the spot where the polar bodies are formed. A single sperma- 

 tozoon appears to effect impregnation. Seginentation occurs 

 while the ovum traverses the Fallopian tube. — On the develop- 

 ment of the Cape species of Peripatus, by Adam Sedgwick, 

 M.A., part 2 (plates 12-14). This part is chiefly devoted to a 

 consideration of the segmentation of the ovum and the formation 

 of the layers. The important and in several cases novel facts 

 brought to notice do not readily admit of being abstracted. The 

 embryo at the gastrula stage and in all the earlier stages of de- 

 velopment is a syncytium : no part of the nucleus or centre of 

 force of the unsegmented ovum enters the clear endoderm 

 masses. The gastrula gut arises from an enlargement and con- 

 fluence of the vacuoles in the centre of the mass. — Studies on 

 earthworms, by W. B. Benham, B.Sc. (plates 15, 16, \iibis). 

 In this series of papers the authors describe a number of earth- 

 worms from various parts of the world, placed at his disposal by 

 Prof. Lankester ; these include new genera and species. In 

 this paper, after an historical resume, a notice of the hitherto 

 known genera and species is given, and Microchata rappi, 

 Bedd., is described and figured.— The official refutation of Dr. 

 Robert Koch's theory of cholera and commas. 



The Journal of the Royal Microscopical Society for February 

 1 886 contains :— Fresh-water algoe (including chlorophyllaceous 

 Protophyta) of the English Lake District, with descriptions of 

 twelve new species, by A. W. Bennett, F.L.S. (plates i and 2). 

 A record of a six weeks' collecting in the district between 

 Windermere and Langdale, and no doubt only an indication of 

 what a more protracted study would do for the L.ake flora. — 

 Explanatory notes on a series of slides presented to the Society, 

 illustrating the action of a diamond in ruling lines upon glass, 

 by Prof. VV. A. Rogers. — On the preparation of sections of 

 pumice-stone and other vesicular rocks, by Dr. H. J. Johnston- 

 Lavis. — On the cultivation of Bacteria, by Dr. E. M. Crook- 

 shank (plates 3-5). — On the appearance which some micro- 

 organisms present under different conditions— as exemplified in 

 the microbe of chicken-cholera, by G. F. Dowdeswell, M.A. 



