84 A. A. C. ELIOT MERLIN ON NEW DIATOM STRUCTURE. 



bands of certain Navicula lyra might be possibly "false ghosts," the 

 probable reality of the structure in question seems to be indicated 

 by recent observations on a N. amphisbaena Bory., on the 

 extensive clear central area of which apparently true, fine, very 

 difficult secondaries have been detected. In connection with this 

 subject I may mention that my friend Mr. Nelson lately wrote 

 to me that, with a solid axial illuminating cone of N.A. 1"3 from 

 an oil-immersion condenser, he has caught a glimpse of a similar 

 veil over the valve of Navicula rhomboides. This veil has since 

 been seen by me on other specimens of JV. rhomboides, with a 

 cone of only about N.A. 0"95. 



Melosira westii, W. Smith ; Oamaru. 



This has a fairly regular perforated structure all over the 

 central area. It is more obvious than that on the other species 

 of the genus yet examined, and should be well within the grasp 

 of any good oil-immersion objective. 



Hyalodiscus maximus, Ehrb; Oamaru. 



This diatom exhibits practically identical structure of about 

 similar fineness * on two different focal planes, the valve being 

 apparently made up of two parallel, closely perforated plates. 

 The distance separating the focal planes of these two distinct 

 structural layers is only q-^qq in. 



Two other unnamed species of Hyalodiscus, from the same 

 deposit, show most distinct and unmistakably true double struc- 

 ture of the kind all over the large central area of the discs, the 

 focal planes being respectively y^o anc ^ toVo * n - a P ai 't- These 

 measurements were effected with the fine adiustment of Powell's 

 No. 1 stand and g-in. apochromat by focussing up from the lower 

 to the upper plate of each form, and rejjresent the intervals 

 between the sharply imaged structure on the two layers. 



Of five different specimens of Hyalodiscus mounted on an 

 Oamaru type slide, Moller is only able to give the species of one, 

 so that more precise indications cannot be furnished ; but there 

 seems little doubt that most, if not all, the valves of this genus 

 are similarly built up of two plates. 



* This would not be so in the case of a false diffraction ghost, where, 

 with an over-corrected objective, the upper structure would appear of 

 double the fineness of the lower. 



