16 MORTALITY OF FISHES, WEST COAST OF FLORIDA. 



and translucent; as the transition time approaches, the chlorophyll- 

 like substance responsible for the green color disappears, a red sub- 

 stance takes its place and dissolves in oil globules that have been 

 forming. At this stage it is red and visible to the naked eye. After 

 only a few days the individuals assemble, lose their red color, 

 become encapsulated {Protococcus form), and sink or float on the 

 surface. This stage is followed by a subdivision into two or four 

 new ones. This latter process may be repeated several times until 

 eventually a ciliated form appears again, completing the life cycle. 

 The latter form is covered by an excessively fragile shell which " is 

 broken by the pressure of the thinnest piece of glass." 



This description explains the varying color of the streaks, the 

 periodicity, and it may explain the bursting observed by Darwin, 

 if he used cover glasses in observing them. 



Could this have been the cause of the trouble in Florida? No 

 evidence of the presence, in extraordmary numbers, of these organ- 

 isms was found. When the water was described as red, further 

 questioning brought out the fact that it was by transmitted light 

 on the beach, and not by reflected light. Fishes observed while dying 

 were in clear water. Still, the condition as described may have 

 existed before the writer arrived, and the dissolved decomposition 

 gases might have persisted in the water to an extent fatal to fishes 

 after the organisms responsible had perished. Or the organisms 

 might have become lost from view either by disintegration or by 

 passing into the protococcus form and sinking, as suggested by 

 Carter (1858). But such red water, to explain the conditions found, 

 would have to be distributed over an unprecedentedly large area to 

 persist for two months and yet escape detection by the many passing 

 ships, and it would be necessary to explain the widely fluctuating 

 periodicity, in some cases one year, in two other cases 24 and 25 

 years, respectively. 



Gilchrist (1914) definitely ascribes certain instances of mortality 

 on the South African coast tc Noctiluca and to diatoms. These 

 forms, decaying in large numbers in the water, make the latter an 

 unfit medium for fishes. Among the instances described by him as 

 due to obscure causes, one is clearly due to Peridiniwm. 



DISEASE AND PARASITES. 



;,, 3. Evidence of disease or parasitism is likewise lacking, as is shown 

 by the examination and dissection of specimens already described. 

 On the other nana, it would be difficult to believe, without the most 

 thoroughly convincing evidence, that so many species of animals 

 could be affected by an epidemic of any single disease, or that such 



