NATURAL HISTORY OF THE DIATOMACEyE. 425 



constitute the top and bottom of the box. Besides this colored matter, 

 which gives the peculiar tint to the diatoms when seen in mass at certain 

 times, but apparently not at all periods of their existence, we find the 

 diatom individuals to have scattered throughout their contents, but 

 always within the layer of endochrome, one or more clear globules look 

 ing like oil. These have been called the &quot;oil globules,&quot; and, when the 

 diatomaceae were ranked in the vegetable kingdom, were supposed to be 

 the representatives of the starch found in the larger plants. Ehrenberg, 

 who did (and, in fact, does still) rank the diatomaceae among animals, con 

 sidered these clear spaces to be stomachs, and fancied he had been able 

 to feed these organisms on such colored matter as indigo, and see it enter 

 these spaces. On account of their frequently being present in numbers, 

 he constructed for the diatoms the name already spoken of, Polygastrica, 

 or many-stomached. There is no doubt that diatomaceae will absorb 

 indigo, or similar material, along with water, and thus their cell-contents 

 may become tinted ; but such taking into their interior, of matter, does 

 not prove their animal nature, as all the Protista absorb solid nutriment, 

 and, in fact, many undoubted plants, under certain conditions, do the same. 

 The typical form of a diatomaceous individual, then, consists of a little 

 siliceous box with its cell-contents more or less colored of an olive-brown 

 tint. The variation in form of the top and bottom of the box, which are 

 known as the &quot;valves,&quot; is very great, whilst the band uniting the valves, 

 and called the &quot;connecting membrane,&quot; or &quot;cingulum,&quot; remains essen 

 tially the same, being merely a ring, narrower or wider as the case may 

 be, and conforming in contour, of course, to the valve to which it is 

 attached, and whose outline it then typifies. In the same way, the sides 

 of a trunk form an oblong ring, and those of a pill-box a circular or oval 

 loop. Thus, in the genus Pinnularia, the connecting membrane is 

 oblong, with rounded ends ; in Coscinodiscus, and the many similar 

 genera, circular; in Triceratium, triangular; in AmpJiitetras, quadrangu 

 lar, and so on, as can readily be understood, and will be made more plain 

 when the various forms come to be seen in their integrity. As has been 

 said, the outline of the valves varies very greatly, but is, for the most 

 part, constant in each genus. We occasionally find, however, that the 

 outline of the valve is the same in two or more distinct genera, which 



are then separable by means of some other character. The intimate 

 VOL. i. 56 



