1892.] 



Development of the Stigmata in Ascidians. 



511 



of Botryllus ; and there can be, accordingly, just as little doubt that 

 the development presented by the oozooids of these forms is also 

 Secondarily abbreviated. This conclusion is corroborated by the 

 facts that in these genera and their allies the ova contain more food- 

 yolk than is usual in Ascidians, and that the duration of the free- 

 swimming larval stage is greatly reduced. 



It is therefore obvious that the primitive (phylogenetic) mode of 

 development of the stigmata in Ascidians is by the subdivision of 

 transverse protostigmata arranged in a single longitudinal series on 

 each side of the pharynx. 



I have observed this process in two other species of Ascidians, 

 Thylacium sylvani (Carus)* and Styela (Styelojpsis) grossularia (van 

 Beneden). These two species are very closely allied, and exhibit no 

 appreciable differences in their mode of development. 



In Thylacium sylvani eight protostigmata arise on each side of the 

 pharynx, and become subdivided, in regular order from before back- 

 wards, to form a corresponding number of rows of secondary stigmata. 

 The protostigmata extend right across the sides of the pharynx, as in 

 Botryllus, before they are subdivided ; and, although I have not 

 actually observed their earliest stages, they give every appearance of 

 having been formed in regular order from before backwards. The 

 subdivision of the protostigmata begins towards their dorsal extremi- 

 ties and then extends ventrally — a process which compares very well 

 with the formation of the stigmata in Clavelina. In the pharynx, at 

 an early post-larval stage, and after the subdivision of the protostig- 

 mata has commenced, the actual nature of the process can easily be 

 observed. Small projections arise from the anterior margins of the 

 protostigmata, and are met by corresponding outgrowths from their 

 posterior margins ; the tips of these projections then coalesce with 

 one another, and by their union give rise to the so-called interstig- 

 matic bars of the fully constituted branchial sac. The shape of the 

 secondary stigmata during the process of subdivision is often very 

 irregular, but an admirable symmetry and regularity of form and 

 arrangement is presented as soon as the subdivision is completed. 



The following numbers illustrate the regular order in which the 

 protostigmata are subdivided. They represent the numbers of 

 secondary stigmata into which the protostigmata of one side have 

 become converted in a pharynx 1*125 mm. long and 2*5 mm. in cir- 

 cumference, at a period when the stigmata of the first row have 

 already assumed the form of narrow, longitudinally elongated slits : — 



* I owe to Professor Ray Lankester the opportunity of examining the type- 

 specimen of this species, which is in the collection of the Oxford University 

 Museum. It was vary erroneously described by its discoverer, and a redescription 

 of it, which I have prepared, will be published shortly. 



