TRANSACTIONS OF SECTION D. 717 



could possibly be called by this name are two broad, shallow pouches at the 

 posterior end of the peribranchial sacs, one for each side. i?hey certainly 

 have nothing whatever to do with the budding, since the buds arise about as 

 far away from them as the size of the ascidiozooids will permit. Furthermore, 

 they do not have the same relations as tlie epicardium of Clavelina and other 

 ascidiaus. In Goodsiria and Botryllus, I may add, they are merely parts of 

 the peribranchial sacs ; while in other cases they arise in a definite way irom the 

 branchial sac. 



In uiy opinion it is an unjustifiable and purposeless forcing of things to attempt 

 to see anything in either Goodsiria or Botri/Uus that is homologous with the 

 epicardium of Clavelina and other budding ascidians. 



Relying chiefly on the evidence from adult structure, we are, then, as it seems 

 to me, obliged to conclude that the compound ascidians have arisen from the 

 simple ones by at least two distinct groups of these latter having independently 

 acquired the property of reproduction by budding. Now, since the processes of 

 evolution are of quite as much scientific interest to us as are its products, we can 

 hardly avoid an attempt to gain some insight into the developmental processes 

 that have been in operation in this instance. 



One question we are impelled to ask is whether some cause for the origin of 

 buddinc in these animals may not be detected here, where it, whatever it is, has 

 been so potent as to produce its eflect twice. A possible cause does suggest itself, 

 and I venture to present it to you very briefly. I confess, however, that the 

 venture is made not without some trepidation. 



It will be remembered that we have given reasons for regarding Goodsiria and 

 Ferophora as simplified or pigmy Polycarpse and Ascidite respectively. It seems 

 to me possible that budding might have arisen in these genera of simple ascidians 

 as a result of the diminution in size and simplification in structure of some of the 

 species ; and I am disposed to regard the diminution in size as the most important 

 factor. It appears to me that the smallest species of Poh/cnrpa, for example, have 

 a much poorer chance of survival than do the larger and largest ones, owing to 

 the simple circumstance that they cannot produce anything like so large a number 

 of embryos as do the larger species. The smallest species that I know of this 

 genus is only 3 or 4 millimetres in length, while most of the species reach some 

 centimetres at least in length ; and it is a matter of common observation that in 

 the ascidians the size of the ovary and the number of the ova are in direct propor- 

 tion to the size of the parent individuals. It is certain that the total volume of 

 the sexual products of a large I'olycarpa would be many times greater than the 

 entire animal of the small species to which I have just referred ; and the ova in the 

 one case are not much, if at all, larger in the one than in the other. The suggestion 

 is that in these cases budding has in some xcay arisen as a compensation for the 

 diminished power of sexual reproduction. 



A developmental question of wider moment than the one just disposed of, and 

 one which I discuss with much gi'eater confidence, is this. If blastogenesis has 

 had two or more wholly independent origins among ascidians, how is the close 

 similarity in the development of the blastozooids of the whole group to be 

 explained? The interest of this question is greatly increased by the fact that not 

 only is the development of the blastozooids much alike in all the species, but also 

 that this development is quite unique as compared with the development of the 

 embi-yozooid. 



In contrasting the development from an embryo and from a bud it is seen that 

 in einhri/onic development the ectoderm produces the matrix of the test, the peri- 

 branchial sacs, and the central nercons si/stevi and hypophysial duct, while in the 

 bud we see these four parts of the animal produced by the inner or so-called endo- 

 dermic vesicle. 



Concerning the endodermic, or ratlier inner vesicle origin of the ganglion and 

 hypophysial duct, I speak with perfect confidence as regards Goodsiria and 

 Perophora, for this confidence rests on my own observations. The case for the 

 Goodsiria bud in particular is as clear as anyone could wish a developmental 

 fact to be. 



