554 V. W. GAMBLE. 



A continuous air-curreiit was led into the water, and the 

 coloured screen was cut so that its halves embraced the air- 

 tube, which was blackened at this point. The junctions of 

 the screen with the bell-jars consisted of black velveteen so 

 as to cut out any oblique white rays, but it was found that 

 great care is needed to avoid liquefaction of the gelatine films. 

 A trial was made with Schott's coloured glass, but except the 

 red the samples submitted were not monochromatic. 



In order to observe the prolonged effect of monochromatic 

 lijj'ht, and to obviate the dominant inlluence of the back- 

 ground, fluid screens were constructed. To insure a fairly 

 strong light the screen was made of one cell only, and not, as 

 in the case of Landolt's original design, of two or more. A 

 double glass vessel was employed consisting of two beakers or 

 of two large cuvettes, the inner one standing on glass supports, 

 so that its rim just cleared that of the outer vessel. The inner 

 vessel was then provided with young, transparent Hi ppoly te 

 in filtered water, and finely divided Ceramium was used as 

 food. The space between the two was then tilled with the 

 colour filter until the level exceeded that of the water in the 

 inner vessel, the top inch or so of which was rendei-ed opaque. 

 A cover of glass, or of glass and gelatine, was then placed 

 over the double vessel, and the whole was then transferred to 

 a shallow aquarium in a strong light. In one case a circula- 

 tion of tank-water was maintained in the inner vessel. The 

 main point of the apparatus is to provide a means of flooding 

 the animals (which remain in mid-water attached to their 

 weed) with transmitted coloured light, and thus largely to 

 iivoid tiie affect of light reflected from an absorbent or 

 reflecting background, such as has been generally employed 

 in previous experiments. The surfaces on which the vessels 

 stood were either slate or dull white brick, but there was 

 always a layer of the fluid, some 2 cm. in thickness, btween 

 the bottoms as well as between the sides of the two vessels. 



The coloured solutions employed consisted of the following : 

 For i-ed a strong solution of erythrosin in distilled water, the 

 strength being increased until a 2 cm. layer cuts out all the 



