336 PROCEEDINGS OF THE ACADEMY OF [Dec.^ 



different refractive indices, but its appearance under high powers 

 indicates that the tubules are not penetrated by the mounting 

 medium. Its most important characteristic for the present purpose 

 is that the markings do not extend throughout the valve but are 

 confined to curved bands with clear interspaces and a large area of 

 clear silica in the center. On .examination with polarized light, 

 using the highest power and aperture objective that will stand dark 

 field illumination without halation, which in my experience, is a 

 4 mm. of .65 N. A., it will be noted that the polarizing effect is 

 confined strictly to the tu}:)ular markings, while the interspaces and 

 central area remain dark, indicating that the silica, itself, is entirely 

 devoid of polarizing action. 



Probably everyone who has mounted many slides of selected and 

 arranged diatoms, has at times been annoyed to find that the gelatine 

 fixative has sealed up some of the cellular structure of the diatoms, 

 preventing penetration by the mounting medium and producing 

 unsightly patches. However unsightly, they become interesting 

 with polarized light as the manner in which the air filled cells light 

 up, while the remainder of the valve is practically invisible, can 

 leave no doubt in the mind of the observer as to the cause of the 

 polarization effect. This was particularly noticeable on an arranged 

 group of marine diatoms from California, which on completion, I 

 regarded as greatly impaired by the manner in which the central 

 portion of several valves of Hyalodiscus maximus had been rendered 

 semi-opaque by the gelatine preventing complete penetration of 

 the balsam. With polarized light, the minute air filled c^Us closely 

 resembled, except in their symmetrical arrangement, the correspond- 

 ing assumedly air filled cells of the secondary structure of Aulacodiscus 

 formosus. A similar condition seems to exist naturally in Hyalodiscus 

 suhtUis, in which the so called umbilicus contains markings like the 

 margin, but being impenetrable to balsam, remains comparatively 

 dark when mounted in that medium, while the outer zone becomes 

 very transparent. It is hardly necessary to add, that the central 

 space reacts with polarized light. 



Microscopists accustomed to the employment of polarized light 

 for the purpose of determining whether a substance is isotropic or 

 doubly refracting, have long been familiar with the effects of reflec- 

 tion and refraction and the necessity of avoiding them by the use of 

 parallel illumination and mounting media of similar refraction to 

 the material examined. In this explanation. I have confined myself 

 to phenomena connected with the effect of •^uch reflection and 



