26 JOURNAL OF THE [January,. 



bosses, of pittings, and of perforations. Until very recently the 

 preponderance of opinion has been with the advocates of the 

 existence of hemispherical protuberances from the outer surface of 

 the valve. The valve itself has generally been taken to be a 

 single layer of silex. But of late a good deal of evidence has 

 accumulated to show that it is composed of at least two layers 

 and perhaps three, and the idea is rapidly gaining ground that 

 the dots are perforations of the middle layer, if there are three 

 layers, or of the inner one, if there are only two. The outer- 

 most lamina and the innermost also, if the threefold theory holds, 

 are supposed to be exceedingly thin, the main weight of material 

 being in the perforated layer between. In fact, the outer layers 

 i.re regarded as mere membranes overlying a sieve-like wall, of 

 v.^hich the areolse may be round, square, or hexagonal in form. 



This last-named theory accords best of all with our belief in 

 the vegetable nature of the diatoms and what we know to be the 

 requirements of the vegetable cell. The sum and substance of 

 this theory is that the cell-wall is not a solid and impervious mass 

 of silex, as was formerly supposed, but that the arrangement of 

 layers which I have described gives it a semi-punctate structure 

 which, while affording all necessary strength, allows full play to 

 the vegetative processes which, through osmotic action or other- 

 wise, depend upon communication between the enclosed endo- 

 chrome and the exterior world. 



But all this brings us once more face to face with the biolo- 

 gical problems with which we began the consideration of this 

 subject this evening. As I predicted at the beginning of this 

 lecture, our latest and most extended knowledge, like our earliest 

 and simplest, ends at the everlasting interrogation-mark. There 

 never can be a finality to human research. Physicists speak with 

 some confidence of an ultimate indivisible unit, the size of which 

 they even undertake to estimate in a rough sort of way; and yet 

 if, in the course of ages to come, the power of the microscope 

 should actually reach that degree of development which was 

 falsely claimed for it a century and a half ago (when Joseph 

 Highmore and others declared that the lenses of that day 

 enabled one to see "the atoms of Epicurus" and "the subtile 

 matter of Des Cartes"), I have no idea that there would be a 

 cessation of microscopical endeavor, or that there would be any 



