OrronKK. 1^12. 



KNOWLEDGE. 



371 



in this form the same structurolcss mcml)rane, 

 stretched upon the to[) to cover the sides of 

 the hexagonal framework, as in T. faviis. In 

 Figure 407 

 it is seen span- 

 ning a gap. 

 and half-wax 

 across one of 

 the character- 

 istic circles of 

 white spots. 

 Note, also, the 

 outer sides of 

 the rings 

 \\here the 

 membrane is 

 torn awa\,and 

 see how 

 exactly they 

 agree, both in 

 this and in 

 Figure 408, 

 with similar 

 St r u c t u r e 

 s h o w n i n 

 Figures 392, 

 393, 395, 396, 



and 397. Can anyone doubt, also, that in Figure 

 405 we see the same disposition of structure as 

 in I-'igure .i ? 



In the torn structure of T. favus,\W- write-r thinks. 



Figure 407. Cuscinodiscus astcromphaliis. Torn slnicUut- imm Himiher valve. 



Objective, Baker's oil-iniiiiersion, ordinary plate, no screen, magnification two thonsand 



seven hundred and fifty diameters, exposure twelve mimites. 



Figure 40S. From the same valve as Figure 407. 



the compound triangular and discoid forms are con- 

 structed are the same. The hexagonal framework 

 and the under la\"er of eve-spots can be demonstrated 



h \- direct 

 [) r o o f — is 

 analogy to 

 stop short at 

 this point be- 

 cause the liner 

 structure at 

 the t o p , 

 though seiMl, 

 cannot he so 

 dealt with? 

 W li e n a n \ 

 detail o"f 

 structure is no 

 linger within 

 the grasp of 

 the aperture 

 of the objec- 

 tive, it is onl\- 

 imperfec t ly 

 resolved. Dr. 

 .\bbe, more- 

 over, has told 

 us in his 

 paper, " On the relation of aperture to power," that 

 under these conditions, all such details, whether 

 triangular, square or lozenge shaped, assumt; the 

 same rounded or oval outlines. In P. formosum. for 



Figure 409. Coscinodiscus centralis. Objective, 



Baker's oilinmiersion, ordinary plate, pot green 



screen, magnification two thousand seven hundred 



and fifty diameters, exposure three hours. 



lies the key to secondary structure generalK . One 

 admits that when we come to the finer it must be 

 more from analog}" than from direct proof. If, on 

 the other hand, we reject the inference we are met 

 with this difficulty. The general lines upon which 



Figure 410. Anlacodiscus Kittonii. Objective, 

 Swift & Son's one-twelfth inch oil immersion of 

 1-30 N..A.., orthochromatic plate, yellow screen, 

 magnification one thousand nine hundred 

 diameters, exposure twenty minutes. 



instance, the " heads " were always spoken of as 

 round, \et under the 1-40 apochromatic in the 

 writer's print of the broken valve with the postage- 

 stamp fracture they come out distinctly square. 

 Figure 409 is another specimen of the discoid 



