520 SUMMAKY OF OUKRENT RKSEAKCHES RELATING TO 



set up in the mantle-cavity ; the pectinate gill forms a water-pump and 

 a food-sieve ; the fine particles arrested in the inhalant chamber reach the 

 mouth by a food-groove on the right side of the body, the coarse particles 

 by way of a food-pouch placed in front of the mouth. The radula is 

 used for grasping the food-masses and conveying them to the mouth. 



The main food-current is caused chiefly by rows of ciha, the lateral 

 cilia, on the anterior and posterior faces of the gill-filaments. The food- 

 streams are caused by rows of cilia on the dorsal and ventral faces of the 

 gills, by cilia on the dorsal surface of the animal, and by cilia on the 

 inside of the mantle. Orton discusses the analogous food-streams in 

 Lamellibranchs. The mode of feeding in Calyptraea chinensis and 

 Cafulus Imngarims is the same as in GrepicMa, and there is therefore 

 reason to suspect that all sedentary Pectinibranehs feed in the same or 

 in a similar manner. 



Spermatog'enesis of Snail.*— Arthur Bolles Lee continues his in- 

 timate studies on this subject, dealing in this communication with the 

 strepsinematic stage of the male auxocytes. 



Structure and Origin of Pearls.f— H. Lyster Jameson has made a 

 critical study of this question, and comes to the following conclusions : 



1. The evidence that the globular Cestode larva, which Herdman 

 regards as the cause of the formation of " fine pearls " in the Ceylon 

 Pearl-oyster, are a young stage of the worm described by Shipley and 

 Hornell as Tetrarhynchus unionif actor, is quite inconclusive. They are 

 more probably referal)le to the genus Tylocephahim. 



2. The theory that these tapeworms are the cause or a cause of the 

 formation of pearls in the Ceylon Pearl-oyster (in the sense in which of 

 the Trematode is the " cause " of pearls in Ilytilvs, where the pearl-sac is 

 normally formed as a result of the specific stimulation of the worm) is 

 supported by quite insufficient evidence, and even their occasional 

 occurrence in the nuclei of Ceylon pearls has yet to be demonstrated. 



3. The shell of Margaritifera has. like other typical bivalves, a perio- 

 stracum, prismatic substance, nacre, and hypostracum or muscle attach- 

 ment substance. In addition to these, certain pathological varieties of 

 shell-substance arise when the normal rhythm of secretion is disturbed, 

 the chief of which are "amorphous repair substance" (probably uncalcified 

 conchiolin), " columnar repair substance," and " granular repair nacre." 

 These substances intergrade with normal nacre and prismatic substance, 

 and with each other. The peculiar characters of these substances are 

 the chief cause of the distinctive appearance of the " pseudo-nuclei " of 

 pearls. The shell substance, except the hypostracum and the outer layer 

 of the periostracum (and probably the ligament) is secreted in liquid 

 form, and its structure and variation may be interpreted as the expres- 

 sion of the normal processes of the crystallization of CaCOg in a colloidal 

 medium, modified by the periodicity of the action of the shell-secreting 

 tissues of the mantle. ' 



4. The " calcospherules," which Herdman identifies as the nuclei of 



* La Cellule, xxviii. (1912) pp. 221-53 (1 pL). 



t Proc. Zool. Soc. (1912) pp. 260-358 (15 pis. and 9 figs.). 



