MEMORANDA. fill 



a species is defined, and then we get /8, y, 3, &c., noticed a$ 

 varieties, each with characters at variance with the character 

 of the species. A genus must be characterized so as to 

 include every species referred to it ; and in the same way a 

 species must have a character that will include all its varieties, 

 or at least not exclude any one of them. No one has a right 

 to say which variety is the type of a species, when all may 

 have arisen from the same original seed ; the primitive form 

 may be y8, or t, or \, of that we are totally ignorant. All we 

 can do is to arrange the varieties according to some arbitrary 

 rule, and to define each, a as well as 0, or tt ; the peculiar 

 character of a cannot be incorporated with the diagnosis of 

 the species without doing violence to the other varieties, and 

 in fact separating them as distinct species, although the author 

 lias not had the courage to do so, from indeed feeling con- 

 vinced they are not so. 



Although I consider the distance between the striae to 

 depend considerably on the age and size of the frustule, I 

 wish it to be distinctly understood that as yet we have not 

 sufficient information whether the number of striae may not 

 also vary slightly in the same way. It is possible that both 

 variations take place ; but until the law connecting them be 

 ascertained, all attempts to derive specific characters from the 

 striae are futile, if not injurious, to the science, unless we 

 allow a very considerable range, so great indeed as to destroy 

 the utility of such characters. If, as some microscopists 

 think, the striae are the walls, and dots the angles formed by 

 the walls of minute compressed cellules, the surface of the 

 valve of a diatom ought to be compared with the cuticle of a 

 flowering-plant, which consists of cellular tissue so highly 

 compressed that the upper and lower walls touch each other, 

 leaving the lateral ones in the form of reticulating veins, these 

 vein-like reticulations having a certain thickness as well as 

 height : in viewing these by direct light, Ave see only the 

 breadth of each line, while obliquely, or by oblique light, 

 causing an oblique shadow or picture to enter the object-glass, 

 we see also its height, and therefore a greater surface of the 

 reticulating lines I'eaching the eye by oblique than by direct 

 light, we see these lines more distinctly. If the striae of 

 diatoms be formed in that way, we might easily understand 

 that the ease of making them out does not depend on the 

 distance from each other so much as on the breadth of the 

 lines or dots in connexion with their elevation ; and therefore, 

 although the actual distance of such be the same, we may 

 resolve some frustules of the same species more easily than 

 others. Diatomacese are thus not so good tests for the micro- 



