Duggar — Relation of Marine Algae to Salt Sohiti(yis. 479 



any case, algae selected should be sach as would lend them- 

 selves to easy observation under the low power of the micro- 

 scope. Moreover, among red forms it is well to select those 

 which, on being injured or killed, are from their own color- 

 ing matter stained diffusely yellowish, pinkish or purplish. 

 The latter phenomenon greatly facilitates observation, al- 

 though it may lead to error if used as the only indicator 

 where salts of the heavy metals are employed. Algae with 

 long cells are in most cases to be preferred. 



In both laboratories the same general method of arranging 

 the experiments was adopted. The solutions employed were 

 placed in slass cylinder bottles each from 25-40cc. capacity; 

 the cylinders were arranged serially in granite iron crates ; 

 and the latter were set in a shallow aquarium with running 

 sea water (the water being almost to the level of the liquid 

 in the tubes). By such means a uniform temperature was 

 maintained, which condition is of the first importance. The 

 experiments were all conducted in diffuse light. 



The algae used were obtained in the best conditionpossible — 

 in a few cases collected always from the same locality — and 

 these were kept in the labaratory aquaria twenty-four hours 

 before being used. Numerous control experiments accom- 

 panied every series, and in any case where the plants in the 

 control experiments became unhealthy, the entire series was 

 discarded. Owing to the size of the vessels which it seemed 

 desirable to employ, Grinnellia and Nitophyllum, the two 

 membranaceous algae, were cut into fragments about two 

 sq. c. m. in length; while in the cases of the others only the 

 terminal portions of the thallus branches were employed. 

 Nevertheless, under the best of conditions some injuries will 

 occasionally result to the parts of the plants used, and these 

 injuries may not be detected in time, so that slight discrep- 

 ancies in the results are not to be wholly obviated. 



The concentrations employed for each alga were governed 

 largely by a preliminary test. There were commonly from 

 six to eight dilutions of each salt in the final test. These did 

 not prove sufficient in all cases to include on the one hand a 



