230 SCIENCE PROGRESS. 



which we may compare directly with the so-called pseudo- 

 gastrula of Sycon. If this comparison be valid, then the 

 free-swimming larva is not to be regarded as a blastula ; 

 this stage is passed over in the maternal tissues, and is to 

 be seen in figs. 20a and 18 of Sollas's plate xxxvii. The 

 blastula proper is followed by an invagination, as we see 

 clearly in figs. 33, 34, 19, 27, etc., in spite of the distortion 

 that Sollas's specimens seem to have undergone. We must 

 next suppose that at the time of leaving the mother the em- 

 bryo again assumes a blastula form, just as is known to occur 

 in Sycon. Then the free-swimming larva would exactly 

 represent a modified amphi blastula. Whether these assump- 

 tions are justified or not, it is at least clear from Sollas's 

 figures that an invagination does go on in the maternal 

 tissues. 



We have now discussed the chief types of sponge 

 development, and starting from the well-ascertained facts 

 of the development most usual both in calcareous and 

 siliceous sponges, certain assumptions have been made 

 with regard to the course of development in the more 

 primitive forms, such as the Ascons, Halisarca, Oscare/Ia, 

 and Plakina. How far can we now construct a type or 

 fundamental plan of development for sponges in general ? 



Taking into consideration all the modes of develop- 

 ment as yet known, the following is perhaps the most 

 primitive type. The egg by segmentation gives rise to a 

 hollow blastula composed of a single row of flagellated 

 cells, which, perhaps, were originally all alike, but in all 

 forms known to us differ to some extent at the two poles, 

 usually markedly so. The cells of the hinder pole become 

 displaced into the interior, by immigration (Ascetta), invagi- 

 nation (Oscare//a) or epibole {Sycon and siliceous sponges 

 generally). We now have a two-layered larva, and so far 

 the development does not differ in any essential point from 

 what we know in other Metazoa, but events which now 

 come about are altogether peculiar to sponges. The inner 

 mass bursts out and envelops the flagellated layer which 

 was before external to it. This is the so-called amphi- 

 blastula, followed by the amphigastrula stage. The 



