CHAP, xi POLYZOA 387 



POLYZOA ECTOPKOCTA 



In the vast majority of Polyzoa Ectoprocta the egg is fertilized 

 whilst it is still iu the maternal tissues, and undergoes the first 

 stages of its development there. It finally emerges as a free- 

 swimming larva, with more or less degenerate gut, and, after a short 

 tree existence, fixes itself and grows into the first person of the future 

 colony. In a few cases, however, the eggs are shed into the sea and 

 are fertilized there, and a comparatively long larval development 

 ensues ; this type of larva has a well-developed gut, and can feed itself. 

 In these latter cases we have obviously the primitive type of Ecto- 

 proctan development, and it is they which deserve our closest attention. 

 They have been most carefully studied by Prouho (1892), and we 

 select for special description one of the forms described by him which 

 has a long larval development. 



MEMBRANIPOEA PILOSA 



Membranipora pilosa is a species occurring abundantly around 

 the coasts of Europe as a delicate lace-like incrustation on the fronds 

 of Laminaria. Closely allied species are found in similar situations 

 all over the world. Mevibranipora is easily kept living iu vessels 

 of clean sea-water ; the eggs are freely discharged, and develop into 

 the young free -swimming larva, which is termed Cyphonautes, a 

 name bestowed on it when it was supposed to be an independent 

 organism. Its true nature was shown by Schneider (1869), who 

 captured it in the sea and watched it metamorphose into Membrani- 

 pora. To rear Cyphouautes in captivity up to this stage would 

 require arrangements for feeding it with diatoms such as have been 

 employed with success in the case of many other larvae. 



If the vessels in which the larvae are kept have been previously 

 coated with a layer of transparent photoxyliu, then, when the larvae 

 fix themselves, they can be removed from the sides of the vessel 

 together with the photoxyliu to which they are adherent, and cut 

 into sections. Kupelwieser (1905), to whom we owe this method, 

 has given us the best account of the metamorphosis of the larva. 

 He paralysed the free-swimming larvae by adding drops of hydro- 

 chlorate of cocaine to the sea- water in which they were swimming ; 

 the larvae were then preserved in Flemming's fluid or a mixture 

 <;f the solution of corrosive sublimate and glacial acetic acid. 



In the development of the egg of Memlranipora the division 

 into blastomeres takes place in an absolutely even and regular 

 manner, and recalls in a good many ways the segmentation of the 

 egg of Poiijgordius. At the 16-cell stage all the blastomeres are 

 equal to one another in size ; the embryo, however, does not form a 

 sphere, but a biconvex lens (Fig. 312), the axis joining the animal 

 and vegetable poles being very much shortened. The blastocoele is 

 excessively narrow. At the 32-cell stage the flattening is still more 



