io6 



HARD I VI C ICE 'S S CI EN CE-GO SSI P. 



parts, and which we meet with nowhere else in nature, 

 and which distinctly defines these forms. 



The only communication between the surrounding 

 fluid and the interior of the diatom exists at the 

 margins of the connectives, and probably in certain 

 cases at their line of junction with the corresponding 

 valves; and always in the form of linear sutures, 

 visible only under the most powerful of our modem 

 microscopes. We shall probably never be able to 

 find pores or openings, properly so called, in the 

 diatomaceous frustule, in spite of the assertions of 

 certain microscopists : this is an illusion. 



Solid matter never penetrates into the interior of 

 the living diatom : they drink, but eat not. 



(Since the above was printed, M. Deby tells me 

 he is not quite certain that the communication be- 

 tween the external fluid and the interior of the diatom 

 is at the margins, it may possibly exist, particularly 

 in those forms possessing a median line at the ter- 

 minal nodules : this will be referred to further on. 

 — F. K.) 



The movement, so curious, active, and conspicuous 

 in certain diatoms (the principle upon which it de- 

 pends is nevertheless obscure), manifests itself along 

 the sutures, as before noticed. It is also true, that if 

 by some cause the movement of a frustule is arrested 

 by some immovable obstacle, the movement is re- 

 versed. 



It is generally admitted, that when once a diatom 

 becomes free, the valves increase neither in length 

 nor breadth. The connectives, however, continue 

 to increase by means of fresh accessions of silica 

 around their free margins, secreted by the primordial 

 utricle : this growth is necessary for deduplication 

 (self-division), as we shall see further on. 



The valves of diatoms are very often composed of 

 several siliceous layers, closely superimposed, and 

 presenting various forms, according to the genera and 

 species, and bearing on their surface the sculpturing 

 that ornaments them, and of which the varieties are 

 very numerous, and the design generally very elegant. 

 The connectives are also often furnished with similar 

 designs, but more simple. 



We will not now enter into the question as to the 

 intimate structure of the markings on the surface of 

 the valves of the diatoms, some of which defy our 

 most powerful objectives. The simplest form of 

 diatom frustule is composed of — 

 1st. A superior valve (fig. 71, F'). 

 2nd. An inferior valve, sometimes differing from 

 the superior (fig. 71, F). 



3rd. A connective adapted to the superior valve 

 (fig. 71, G'G'). 



4th. Generally a connective adapted to the inferior 

 valve (fig. 71, GG). 



One of these connectives slides within the other, 

 like the draws of a telescope, and can be separated. 



We must consider all diatoms as organisms, formed 

 of two distinct materials, one of which is living, viz. 



the protoplasm, the other non-living. The first alone 

 is able to produce the second ; that is to say, the 

 endochrome and the oily corpuscles in the interior ; 

 and exteriorly, the cellulo-siliceous frustule, with the 

 various supplementary appendages which we meet 

 with under the form of stipes, tubes, mucous cushions, 

 gelatinous envelopes, &c. The endochrome of 

 diatoms, like the pigment found in animals, or 

 the chlorophyl in plants, has not the power of 

 secreting the secondary products ; being themselves 

 formed materials, their role is in connection with the 

 functions of respiration. 



Deduplication (self-division). — If a diatom is ex- 

 amined when the connective is turned towards the 

 observer, we perceive that it increases longitudinally 

 by accretions at the margins of the connectives ; at 

 the same time, at the extremities of the frustule, a 

 fold may be seen in the primordial utricle. (This fold 

 is not at the extremity of the frustule, but at the ex- 

 tremity of the connective of the inferior valve. — F.K. ) 

 This fold gradually approaches the central cavity, and 

 is prolonged by degrees until it reaches the nucleus, 

 which it divides in two, if it has not done so before. 

 The opposing edges of the fold continue to approach 

 each other, and at last unite, and the point of 

 contact is obliterated at the place where we originally 

 saw the nucleus, and which is completely separated 

 (fig. 72). At this time there exist in the interior of 

 the diatom two contiguous primordial utricles, each 

 possessing a nucleus, which quickly becomes central. 

 In each nucleus may be seen a nucleolus. As soon 

 as this phenomenon is accomplished, both external 

 surfaces of the new utricles commence secreting a 

 siliceous covering, which quickly thickens, and is 

 covered with the design proper to the species, and 

 speedily assumes the form and aspect of the old 

 valves (fig. 73). The new interior valves occupy a 

 central position in the frustule, and are face to 

 face. 



We have seen a diatom formed of four valves, when 

 the two old external valves and the two new internal 

 ones were close together ; the new valves were sup- 

 ported all round the circumference by the connectives 

 of the original valves. 



Soon after — sometimes even before — the parting of 

 the primordial utricle, we perceive that the connec- 

 tives are considerably enlarged, and at the same time 

 the interior slides in the interior, and by this means 

 separating the two valves to a considerable extent ; 

 thus augmenting the dimensions of the internal 

 cavity of the frustule. The connectives of the young 

 valves are not developed until later on, — sometimes 

 before their liberation, and sometimes after, according 

 to the genera and species of the diatom. 



Later on, the sliding of the connectives in the 

 species whose frustules live isolated, attains its 

 maximum ; the narrowest liberating itself entirely 

 from that which had served it as a sheath. 



From what we have just said, we may meet with 



