NA TURE 



[Jan. i 5) 1885 



reptans, the test is penetrated in all directions by a well- 

 developed system of tubes with large and numerous 

 terminal bulbs. A series of Simple Ascidians could be 

 formed showing all conditions between these two ex- 

 tremes, and also exhibiting very varied arrangements in 

 regard to the disposal of the vessels in the test, their 

 modes of branching, and the relative numbers and sizes 

 of the terminal bulbs. But perhaps the most interesting 

 modifications of all are those met with in some of the 

 members of the remarkable deep-sea genus Culeolus, 

 There we find a great development of the vessels and 

 their enlargements just on the surface of the test, and 

 separated from the surrounding medium by a very thin 

 layer of tissue. When describing this system in 1882,' I 

 suggested that in these species it might act as an acces- 

 sory organ of respiration, and I have lately shown " that 

 an investigation into the condition of the corresponding 

 system of vessels in some of the Compound Ascidians 

 supports this idea, the chief arguments in favour of which 

 are 1 — 



(1) The disposition of the tubes and cavities in the 

 different regions and layers of the test, and the anatomical 

 characters of the system. 



(2) The relation which exists in many groups of Asci- 

 dians between the branchial sac (the chief organ of 

 respiration) and the system under discussion, — where the 

 branchial sac is large and highly developed, the vessels 

 in the test are few and small ; but where the branchial sac 

 is small, simple, and apparently inefficient, the vessels in 

 the test are numerous, of large size, and disposed in such 



Fig. 4.— Corella jarallelo'ramma. The posterior part of the left side of 

 the test of a specimen from Loch Fyne. Twice the natural size, t'., the 

 system of vessels. 



a manner as to suggest that they are concerned in the 

 aeration of the blood. 



It is obvious that it would be advantageous toanAscidian 

 if its test could act even to a slight degree as an accessory 

 respiratory organ, by allowing the blood circulating in its 

 superficial layers to be brought into such close relation 

 with the external medium as to render possible a certain 

 amount of oxidation. And consequently it is easy to ima- 

 gine the process of evolution of such a complicated system 

 as we find in Culeolus murrayi from a few simple vessels 

 like those in the test of Corella parallelogramma (Fig. 4). 

 But it is probable that the common ancestor of Simple 

 and Compound Ascidians had no blood-spaces in its test. 

 There are none in the " Haus" of the Appendiculariidae ; 

 and in Clavelina, which may be regarded as nearer to the 

 first Simple Ascidian than any other form known, there 

 are no vessels in the test except those of the stolon. Some 

 structure must therefore be looked for from which the 

 first respiratory blood-system of the test may have been 

 evolved, and such a structure is to be found, I believe, in 

 the gemmiparous stolon of the Clavelinidae. 



Clavelina (Fig. 1), which from other independent 

 evidence I regard as the most primitive form of Simple 

 Ascidian known to science, is one of the so-called 

 " Social " Ascidians in which the members of the colony 

 are united by a creeping stolon containing " vessels " 

 (that is a prolongation of the ectoderm and the mantle, 

 and a blood-tube) which place the circulatory systems of 



1 " Zoological Reports of the Challenger Expedition," Part xvii. pp. qo 

 and 279. 



* Prec. Lit. and Phil. Sec., Liverpool, session 1884-85. 



the various members in communication, and from the 

 ends of which, in prolongations of the test (as at B, Fig. i), 

 new members are produced by gemmation. It is possible 

 that this system may act in some slight degree as a 

 respiratory organ, but its chief function, and probably its 

 only one, is the asexual production of new individuals. 



The ancestors of the remaining Simple Ascidians 

 diverged from the ancestors of the Clavelinidae, and lost 

 the power of reproducing by gemmation, but in many of 

 the least modified of the Ascidiidae we still find processes 

 from the posterior end of the test which contain vessels, 

 and so closely resemble the stolon of Clavelina in all 

 particulars that there can be no doubt that they are per- 

 sistent rudiments of that structure. 



In Ciona, which is certainly one of the most primitive 

 of the Ascidiidae, vessels are only present in the posterior 

 part of the test, and here we frequently find them drawn 

 out into long processes of the test, which have the greatest 

 possible resemblance to stolons (Fig. 2), and are doubtless 

 their homologues, although they no longer function as bud- 

 producing organs. They are useful as adhering organs, 

 and they have probably to a slight extent commenced to 

 perform a respiratory function. 



Fig. 5. — Vessels in the surface layer of the test of Ascidia mannnulata as 

 seen in a section magnified about 40 diameters, b, small part of the system 

 more highly magnified; a.v., afferent vessel; t.v., efferent vessel: 

 t.k. t terminal knob; f., surface of the test. 



I imagine then the first stages in the evolution of the 

 " respiratory" vessels to be as follows : — As the ancestors 1 

 of the Ascidiidae lost the power of reproducing by gem- 

 mation, the vascular stolons became rudimentary, until 

 they were useful merely as adhering organs. For some 

 time they would only be produced at the posterior end 

 of the test (their original position in the Clavelinidaj), 

 but in course of time they would extend further forwards 

 along the left side of the body (the side upon which most 

 Simple Ascidians lie) so as to anchor the animal more 

 securely, and we even find them occasionally in this 

 condition in Ciona intestinalis and in Ascidia aspersa 

 (Fig. 3)- 



They would then probably (in some not very remote 

 ancestor of Ciona) begin, while still acting as adher- 

 ing organs, to be of some slight use in respiration, and 

 would, consequently, by the action of natural selection, 

 be evolved gradually into a larger system of vessels 

 extending over a wider area of the test. And here 

 might be shown a series of the Ascidiidae passing from 

 Ciona (Fig. 2) through Corella (Fig. 4), and Ascidia 

 plebeia, in which the system is still very feebly developed 

 and confined to the posterior half of the left side of the 



1 See phylogenetic table in '' Challenger Reports," part xv p 



