WOODWAKD : ANATOMY OF PTEROCERA. 149 



analogous structures described by Yanstone' in Melongena, tbat are 

 too restricted in their area to be solely protective. I further consider 

 that the style is composed of a reserve supply of food material, 

 although it may also function in the manner suggested by Barrois. 

 One of the latter's arguments against the reserve-food theory is that 

 in its reactions the style most nearly resembles mucin ; it, however, 

 differs from it in one important respect, viz., in giving the proteid 

 reaction with copper sulphate, and, as so many of these very different 

 proteid compounds only differ slightly in their chemical tests, it may 

 really be a very different body ; it is, moreover, readily dissolved by 

 the digestive juices, so that it can very easily be absorbed by the 

 cells of the alimentary canal. 



I have, moreover, tried Hascloff's starving experiments, and came 

 to conclusions very different from those of Barrois. Having 

 obtained 100 mussels, and, after examining a number and finding 

 that they possessed well-developed styles, I carefully cleaned 50 

 and placed them in aerated filtered sea-water, the remainder being 

 placed in a feeding tank for comparison. The starved specimens 

 were kept in filtered water which was frequently changed, and 

 one or two were killed every other day for a space of one month ; 

 when examining a starved mussel one of the reserve fat stock 

 was always killed for comparison. After six days' starvation the 

 difference in the size of the style was as much as three-quarters 

 of an inch in two specimens of equal size, that of the starved 

 one being of course the smaller. Examining them in this way, as 

 time went on the difference became more and more marked, till at 

 the end of the month many of the starved specimens had entirely lost 

 their styles. Thus, amongst fifteen starved mussels five showed no 

 trace of a style, five had styles averaging -Sin., while five still 

 retained styles of about lin. in length, but these were mostly very 

 thin ; whereas amongst the non-starved specimens the style averaged 

 about I'Sin. Unfortunately, I had no time to continue the experi- 

 ments and feed up the starved specimens. But I think there can 

 be no doubt that this structure does disappear when the animal is 

 starved, and since it is an albuminous proteid, although differing 

 slightly from all known food-stuffs, I see no reason to doubt its. 

 being a reserve supply of food, which, in spite of the concentric 

 structure, I should regard as aggregated from the digested food in 

 the gut and not as secreted by a ciliated epithelium.- 



With regard to Barrois's assertion that since the majority of marine 

 molluscs do not hibernate they do not require a reserve supply of 

 food, I would suggest that during the winter, when many molluscs 

 migrate into deeper water, and during the breeding season,^ they 



1 Vanstone, Linu. Soc. Jour. Zool. xxiv. p. 369. 



2 These observations were carried out in the Laboratory of the Marine Biological 

 Association at Plymouth, when occupying the table belonging to the British 

 Association. 



3 It is interesting to note that during the process of starvation the mussels all 

 ripened their gonads. 



