1913.] 



Protostigmata in Ascidians. 



451 



transverse rows of stigmata are formed. These are the definitive stigmata. 

 Not only have the protostigmata divided into definitive stigmata before 

 perforation, but also the doubling of the number of rows by a parallel 

 division of all the stigmata in each row has been nearly completed before 

 perforation. Only one stigma of each row (the first one to appear) is still 

 undivided at perforation. A species of Sycozoa [Colella] that I have been 

 able to examine shows a slightly more advanced condition. Usually the first 

 stigma of each row to appear is nearly or altogether divided before perfora- 

 tion, the apertures of the stigmata resulting from the division not connecting 

 with each other. 



Daumezon (1909) states that in Cystodites the process seems to be the 

 same as in Holozoa, but that in Polycitor (Eudistoma) the second of the two 

 first stigmata to appear does not divide, and as a result only three rows are 

 formed instead of four. The adult and larva of his species show an 

 imperforate area in the posterior quarter of the pharynx corresponding to 

 the absent fourth row. Tn a Pacific species of Polycitor (Eudistoma) with 

 three rows of stigmata, but, as far as I can see, without any very definite 

 imperforate area, only three rows of stigmata are formed in the larva. The 

 pharyngo-atrial wall projects into the pharynx in the form of a ridge, the 

 summit of which is the dividing line between the two protostigmatic areas. 

 One row of stigmata appears on the dorsal or anterior side of the ridge and 

 two on the ventral or posterior. In this case it is very evidently the first 

 protostigmatic area that has failed to divide. I have not been able to find 

 any connection between any of the stigmata of the two posterior rows. 

 Complete division has evidently taken place some time before perforation. 

 In two species of Polycitor (Polycitor) in the adult blastozoids of which there 

 are many rows of stigmata, two rows appear on each side of the ridge in the 

 larva. All four rows perforate independently of each other. The first stigma 

 of the third row appears a short time before the first of the fourth row, but I 

 have not been able to see any connection between them. This method of 

 origin is shown in the lowest series of stages of the figure. 



In the Didemnidse I have not yet examined larva? of the necessary age to 

 determine whether all the stigmata perforate independently or whether there 

 is some division of the stigmata, as in Holozoa. The stages that I have seen 

 point to an agreement with Polycitor (Polycitor). In Amaroucium, one of the 

 Synoicidse, two rows appear on each side of the pharyngo-atrial ridge, and I 

 have found no evidence of connection between any of the stigmata. Their 

 origin corresponds with the lowest series in the figure. 



The Dictyobranchia and Krikobranchia agree in many points. They differ 

 from the Ptychobranchia in having only one or two primary protostigmata 



