392 OBSERVATIONS ON PLANT PLANKTON. 



the vegetation of the ocean, it will be at once apparent that the 

 small fringe of visible vegetation on the shallow bottoms round 

 coasts cannot in the least degree suffice for the sustenance of the 

 teeming animal life, which not only extends over the surface, but 

 ranges into the depths. This rule must be played almost entirely 

 by the minute plant organisms, diatoms, Protococcacea, Oscillator iecu, 

 Peridiniea, Coccospheres, Rhabdospheres, &c., which inhabit the 

 surface layers down to 80 to 50 fathoms. Such organisms have 

 been recorded from polar, temperate, and equatorial seas, sometimes 

 in vast shoals, discolouring the water, but always found to be pre- 

 sent when suitable apparatus is used for their capture. Their 

 economic importance to the fisheries is therefore of a direct and 

 vital character. We know that the stomachs of HolothuruE, Asci- 

 dians, Salp(F ; oysters, scallops, whelks, and other molluscs ; crabs, 

 lobsters, and other large Crustacea, and even full grown fishes, 

 have formed the happy hunting-grounds of diatomists in search of 

 material ; but direct evidence has been wanting of the use of 

 diatoms as a matter of daily food by animals in the sea. Observing 

 that the excreta of copepoda and other small Crustacea, largely fed 

 upon by fishes, were in many cases tinged with a faint colour like 

 diatomine, I subjected them in many cases to a minute microscopic 

 analysis, with the result that nearly the whole could be resolved 

 into minute fragments of diatom frustules and their chromatophores. 

 In many cases, it was surprising to note that the chromatophores 

 had passed through, almost unchanged in shape, and even retaining 

 faintly their colour. It was, in fact, often possible, from the 

 characteristic shapes of the chromatophores, and the minute finely 

 sculptured fragments of frustules, to recognize them as belonging 

 to diatoms found in the same capture. In a few cases it was 

 possible to detect a whole unbroken diatom cell within the crusta- 

 cean, but almost invariably the diatoms were reduced to fine 

 fragments. Coscinodiscus and Skeletonema were the usual forms 

 found, and these were the prevalent forms in the sea when my 

 observations were being made. 



It appears, then, to be clear that the animals which are them- 

 selves an important constituent of the food of fishes live in turn 

 largely on diatoms. It has always seemed to me very higlily 

 probable that young fishes eat diatoms directly ; and, in order to 

 put this to the test, Dr. Wemyss Fulton sent me some young sand- 

 eels, taken in tow-net, fifteen miles off Aberdeen, on 16th May, 

 1894; some young flat-fish (? plaice), taken ofl" Montrose, 21st May, 

 1894 ; and some very small clupeoid fishes, taken SOth March, 

 1889. They were preserved in spirit. After burning a few of each 

 in platinum crucibles, the ash was examined, and diatoms discovered 

 in every case. Four dift'erent genera, viz. Skeletonema, Eacampia, 

 Melosira, and Chaitoceros, were found in the sand-eels ; Skeletonema 

 and Nitzschia in the flat-fish (in both cases Skeletonema was pre- 

 dominant); and C'oscmo(/i.scMs (abundant) and Melosira (very rare) 

 in the young clupeoids. In each case, complete valves (in Skeleto- 

 nema, chains of several cells) were found, which appears to indicate 

 that they were eaten directly by the fish themselves, and not within 



