274 TROCEEDINGS OF TIIK MAL.VCOLOGICAL SOCIETY, 



muscle, which pulls the radula witli its erectcil teeth back over the 

 surface of the curtilage into the mouth, and tears the teeth through 

 the Hesh of the victim, thus rasping away a portion of it, and 

 carrying the spoil into the buccal cavity, from which it would pass 

 into the oesophagus. 



A com})arison of the buccal mass of NataJina with that of Tcstacella 

 shows certain interesting modifications, obviously due to the different 

 methods of feeding in the two animals. First, Ave notice the almost 

 entire reduction in the latter of the great constrictor muscles of the 

 buccal mass, the outer sheath of that structure being very thin and 

 containing but few longitudinal and transverse muscular fibres. The 

 various internal longitudinal muscles are well developed, the antero- 

 ventral series and the lateral muscles (Figs. 8 and 9) being larger than 

 in Natalina ; the postero-dorsal series is peculiar in the fact that its 

 component muscles perforate the wall of the buccal mass, and are 

 attached to the left body-wall of the slug instead uf to the odontophoral 

 cartilage. 



The radula-sac of Testacella is very short, and hardly extends 

 behind the anti'rior half of the buccal mass ; on the otlier hand, the 

 functional radula extends below the cartilage for a considerable 

 distance backwards, when at rest even further back than the radula- 

 sac does above (Fig. 8). 



The buccal cavity is much larger, and the odoutophore and buccal 

 mass arc capable of much greater evagination than in Katalina. 



Testacella feeds on worms, which it must seize by a very rapid 

 protrusion of the tougaie ; to facilitate this the buccal cavity can be 

 everted, the tongue pushed forwards by the contraction of the intrinsic 

 muscles of the odontophoral cartilage, and the teeth, which ai'e like 

 barbed needles rather than curved hooks, are shot into the prey ; then 

 all the longitudinal muscles come into play and pull, a considerable part 

 of the worm thus hooked into the large buccal cavity and hold it there 

 while its blood is sucked out. This slug does not normally use its 

 radula as a rasping organ after the manner of Natalina, consequently 

 it does not need any great constrictor muscles for pressing the tongue 

 against the body of its prey, but owing to the more active nature of 

 its victim it needs a more rapidly acting and prehensile tongue. 



The oesophagus of Natalina opens dorsally into the buccal cavity 

 near the anterior end of the buccal mass, and just over the odouto- 

 phore the two salivary glands open into it. These glands are a pair 

 of long compact masses {s.g., Figs. 1 and 2) situated at the sides and 

 meeting above the oesophagus, but remaining quite distinct from one 

 another. 



The Condition of the salivary glands is rather variable amongst the 

 Agnatha, N. Trimeni ' possessing only the right gland and duct, 



* I should be inclined to regard the specimen of X. Trimeni described by Pace as 

 abnormal in this respect, for in most Mollusca where the salivary gland is a single 

 mass, that structure owes its origin to a fusion of two distinct glands, always, 

 however, retaining their two ducts, as in Faryijliaiitn {fdc Godwin- Austen, Proc. 

 Malac. Soc, I, 1893, p. 5). 



