116 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



six of the hooks or barbs as they are attached to a gill-raker blade, showing that each 

 of these, again, has a serrated tip. Fig. 4 is drawn to the same scale as the organisms 

 figured in the other plates hereafter described, whence it is evident that the organic 

 materials are easily secured by such an apparatus — the fineness, elasticity, strength, 

 and least possible resistance of which are evident. 



The whole interior of the mouth is always abundantly supplied with a mucous 

 secretion, which aids in the accumulation of the fine food particles as they are strained 

 from the water, and also aids no doubt in conveying them back to the oesophagus. 

 There is also a delicate fold of the mucous membrane running along each gill-arch at 

 the base of the series of gill-rakers, making there a deep channel, but whether there 

 are definite ciliated tracts with this function of conveying solid particles is not as yet 

 known. One always sees, moreover, in examination of these barbs, more or less food 

 material clogged in their meshes, showing again the manner in which it is lodged by 

 them. The cavity of the back part of the pharynx of the fish narrows abruptly to an 

 apex which, guarded by a constrictor muscle, opens directly into the folded stomach 

 of the fish. 



The comparatively empty condition of the menhaden stomach has often been noted, 

 as also the presence of a greater or less quantity of what appears to be a dark greenish 

 or brownish mud, with a variable quantity of copepods and small Crustacea intermixed, 

 although these latter may be almost entirely absent. This apparent mud, however, 

 is made up of the various organic matters which are always suspended in surface 

 waters, and which the fish, by means of the filtering mechanism of gill-rakers just 

 described, has removed from the water in which he has been feeding. This may be 

 demonstrated not only by observing the habits of the fish when living, and by a 

 study of the gill-rakers, but also by collecting through a filter the organic matter sus- 

 pended in a given quantity of surface waters* from the localities where the fish are 

 taken, and then by comparing such a filtrate with the stomach contents of the animal. 

 Studies carried on in this comparative manner show very well how wide a range of 

 microorganic forms contribute to the sustenance of the menhaden, how they are to 

 some extent localized, and finally some of the ways in which they interact upon each 

 other, besides serving as food to the fish. 



One of the richest feeding-grounds of the menhaden studied during the season 

 was in the estuary of the Acushnet River at New Bedford, Mass., as it was here also 

 that the food in the earlier part of the season was of the most complex character. In 

 a mass of food taken July 12 from the stomachs of two dozen fish one could readily 

 find a few small annelids of the genus Nereis, measuring about a half inch in length 

 usually; a few rotifers of the genus Notomata as nearly as they could be identified; 

 quite a wide range of the smaller Crustacea — small amphipods (Gammarus), young 

 schizopod shrimps, a few ostracoda (Evadne), a few Zoea larvae, with quite abundant 

 Nawplim larvae of different species. It may here be stated of the above-named organ- 

 isms, that they are more abundant in fish taken during the night. I have never 



* This was done in the course of this investigation hy means of a filter, such as is used hy Mr. G. W. 

 Rafter, of Rochester, the Board of Health of Massachusetts, and the Laboratory of the Western Division 

 of the Boston Waterworks, etc., made from a large funnel whose tube is closed at the lower end by a 

 coil of fine wire gauze ; upon this plug of gauze rests a ^-inch stratum of fine white sand upon which the 

 organisms collect as the water filters through; from the sand the material is washed out in a given 

 small quantity of water. 



