74 Botanical Society of Edinbui'gh : — 



The number of strice. — lu some species, perhaps in many, this 

 character is by no means constant. In Navicula vurians, I find that in 

 the smaller individuals there are often 24 to 26 strise in 1 -1000th of an 

 inch, while in the larger there are only 14 to 16, and this in indivi- 

 duals of the same type of outline. Smith describes Pinnularia di- 

 vergens With. 1 1 strisein 1-1 000th inch, while I find it more frequently 

 with from 22 to 26 in 1-1 000th inch — the arrangement, which is 

 peculiar, being the same in both. A very striking example occurs 

 in N. elUptica, which, as we have seen, also varies in form. The 

 species, as described by Kiitzing, has very coarse strise, even coarser 

 than appears by any of the figures. But in a variety to which I have 

 directed attention, and which 1 regarded on this account as a distinct 

 species, till I found a gradual transition to the first-named type, the 

 strise are so very much finer, being about three times more numerous, 

 that the aspect of the frustule is totally changed. In comparing 

 examples of the extreme types in regard to striation, I took indivi- 

 duals of equal size, and I found in one very coarse striae ; in the other, 

 striae so fine as not to be easily seen unless the valve was placed in 

 the most favourable position with reference to the light. 



The appearance of the median line and nodules, — In the coarsely 

 striated variety of N. elUptica, there are lines on each side of the 

 median line, forming a double cone, of which the bases meet near the 

 centre. But in the finely striated variety, these lines are parallel to 

 the median line ; only bending outwards round the central nodule. 

 This assists in giving a very different aspect to the two forms, Avhich 

 yet are connected by a graduated chain of transition forms. 



We have then, if w-e consider only the three characters of form or 

 outline, number of striae, and aspect of medial line and nodules, 

 evidence that great variations may occur in any one of them. Nay, 

 in N. elUptica and N. lepida, variations occur in all three together. 

 In such cases as these last, it is difficult to define the species by these 

 characters in the usual way, and we have apparently no resource but 

 to state the fact of the tendency to vary in one or more of these 

 points, as one of the specific characters. In N. vaiians the arrangement 

 of the striae is always the same, as it is also in Piiinularia diveryens, 

 and many others ; but in N. elliptica even this fails, for tlu; striae are 

 highly radiate in the coarsely striated form, and nearly parallel iu 

 that with finer striae. 



Enormous variations in size occur, even in the same type of form. 

 If Pinnularia meguloptera be I'eferred to P. lata, we have a variation 

 in length from about 20 ten-thousandths of an inch to nearly 80. 



The distribution of Diatoms over the world is one of the most 

 remarkable j)oints about their history. Not only do v/e find, if we 

 examine a gathering from any part of the world, that most of the 

 forms are identical with those of our own waters ; but in tracing 

 these minute organisms through the latter to the earlier sedimentary 

 rocks (and it is said that they occur in the lower Silurian strata, the 

 oldest in which any organic remains occur), we find still the greater 

 number of species to be the same as those of the present day. 



Ehrenberg, in his last great work on the distribution of microscopic 



