56 CEYLON PEAEL OYSTER REPORT. 



intestine, with which it communicates by a longitudinal cleft. The function of the 

 crystalline style is still doubtful. Among more modern views, while Mobitjs, 

 Haseloff, and Hazay have argued that it rejjresents a reserve food supply, 

 Barrois, Pelseneer, and others believe it to function as a lubricant to obviate the 

 danger of sharp fragments, taken in with the food, causing damage to the delicate 

 lining of the intestine. The upper end of the style certainly projects into the stomach, 

 and as it wastes, the hinder part is continuously being pushed upwards to compensate 

 for the loss. According to Barrois, sand and shell fragments are invested by the 

 viscous waste of the style, and so made bolus-like are moved along the intestine more 

 freely and without inflicting injury upon the walls. 



Margaritifera vulgaris is, however, capable of exercising a certain degree of 

 selection in feeding, and sand grains are seldom seen in any numbers within the 

 alimentary canal. And yet the ciystalline style is always present in healthy 

 individuals containing a fair amount of food material in the intestine, as, for example, 

 all the individuals examined in Trincomalee harbour in October, 1902, and upon the 

 Pearl Banks during the 1903 fishery. But out of 43 oysters examined at Galle 

 during June, July, and August, while in a state of semi-starvation, having been kept 

 for ten days and upwards in water containing little suitable food matter, five only 

 showed a crystalline style. The alimentary canal of the whole number contained 

 an extremely small amount of food, and the visceral mass was notably shrunken. 



Still it must be remembered that small sand grains and sharp-pointed sponge- 

 spicules and diatoms are here and there to be met with in the stomach contents, and 

 in these cases the suggested gelatinous investment by the style would be useful. 

 Moreover, the cohesion of particles into a bolus capable of traversing the intestine 

 more readily, brought about by the investment, would, no doubt, also be useful. 



A valvular folding of the intestinal ridges gives entrance to the ascending region of 

 the intestine (Int., 2, Plate VI., fig. 1), which, however, before turning on an upward 

 course, curves backwards along the base of the visceral mass to the left of and parallel 

 with the lower or forwardly directed portion of the descending intestine. At the 

 posterior extremity of the ventral surface of the visceral mass the two intestinal 

 divisions intersect, the ascending section crossing to the right. The intestinal loop 

 (Int.lp., Plate VI., fig. 3) thus formed in the floor of the visceral mass is the visceral 

 loop. From the point of intersection the ascending intestine turns sharply upwards, 

 running parallel with and closely adjacent to the upper part of the descending 

 intestine, the course of the latter lying a little forward and to the left. The portion 

 of the ascending intestine forming the second limb of the visceral loop is small in 

 diameter and somewhat compressed. The anterior fold of the descending intestine is 

 continued into it as a somewhat undulating and irregular dorsal ridge dying off 

 midway along. 



At the point where this division of the intestine assumes a dorsal course, an increase 

 takes place in the diameter, concurrent with the appearance of a great longitudinal 



