4 llev. W. Smith on the Diatomacese, with descriptions 



placed in regular squares or in direct opposition to each other. 

 It may be as well to remark in this connection, that in the follow- 

 ing descriptions when the distance between the striae is mentioned, 

 the distance between the beads in the same row may be known 

 by reversing the application of the terms longitudinal and trans- 

 verse : thus when the longitudinal striae are said to be ^y^oth 

 of an inch apart, it follows that this fraction denotes the distance 

 between the beads in a transverse direction and vice versa. 



In all the forms described in this paper, and in the Navicidece 

 generally, we may observe that each valve is traversed by a me- 

 dian line across which the striae do not pass. The centre and 

 extremities of this line are somewhat enlarged, and these enlarge- 

 ments have been regarded by many writers as openings in the 

 siliceous plates. I have never been able to satisfy myself that 

 such openings exist, and am disposed to regard the line itself and 

 its enlargements as peculiarities little connected with the essen- 

 tial structure or functions of the cell. More important in a 

 structural point of view is the form of the connecting membrane, 

 which in Pleurosigma consists of a narrow ring of silex, and 

 which in no period of its growth appears to have any very con- 

 siderable development. The consequence is that the front view 

 of the frustule is uniformly of a linear, or when the convexity of 

 the valves is considerable, of a linear-lanceolate form, while in 

 Navicula and Pinnularia, as the connecting membrane is often 

 more fully developed, the front view of their frustules is 

 frequently oblong or quadrilateral. Too much importance must 

 not, however, be attributed to this aspect of the Diatomaceous 

 frustule, as its form greatly depends upon the stage to which self- 

 division has arrived, and may vary from linear to oblong, or from 

 very narrow to very broadly-lanceolate in the same individuals. 



The reproduction, or more strictly speaking, the multiplication 

 of these organisms, has been noticed only under the form of self- 

 division, the phsenomena connected with which I have detailed 

 in a former paper (Ann. Nat. Hist. 2nd Ser. vol. vii. p. 4) ; I 

 give an example in the present connected with Pleurosigma at- 

 tenuatum (PI. II. fig. 13). By means of self-division the numbers 

 of these minute objects increase with amazing rapidity, and the 

 consequence is that when found they generally occur in multi- 

 tudes so vast that their aggregation forms a brownish film on 

 the surface of the mud, or a visible covering to the stones or 

 plants to which they attach themselves. That these great 

 numbers frequently arise from self-division is evident from the 

 circumstance of their being nearly all of exactly the same size, in 

 the same locality ; and that they may have originated within a 

 brief period, will not appear surprising if we consider the rapidity 

 with which increase in geometrical progression advances even 



