438 C. M. CHILD. 



more than others. Such highly resistant forms as Ceramium 

 rubrum (Child, 'i6a), however, usually show the same gradient 

 in detached specimens as in those collected in situ. In fact the 

 frequency of irregularities and reversals in the gradient and the 

 ease with which they can be induced experimentally constitute 

 in some degree a measure of the sensitiveness of the species to 

 changes in environment. Experiments on Griffiihsia to be de- 

 scribed later will show some of the possibilities in this direction. 

 All of these cases of alteration or reversal of the gradient, what- 

 ever the processes and conditions involved, are of interest in the 

 present connection since they all constitute additional evidence 

 for the existence of a gradient and its fundamental relation to the 

 physiological condition of the plant, and the establishment of 

 these facts is the chief purpose of these studies of algae. 



The visible death changes in the protoplasm of cells stained 

 with neutral red and then killed either with neutral red itself or 

 some other agent consist, as already noted (Child, '160), first, 

 in a deepening of the red color of the dye, indicating increased 

 acidity, followed by an aggregation of the protoplasm into sepa- 

 rate masses which rapidly contract and become black or purple. 

 Apparently during this stage of the process there is an increase 

 in acidity in the cell as indicated by the change in color of the 

 neutral red, but this is followed by a more or less rapid loss of 

 color from the masses of coagulated protoplasm and at least 

 often the cell-contents apparently become alkaline, if the neutral 

 red can be trusted as an indicator. 



These death changes are most striking in elongated cell bodies 

 and are clearly seen in Bryopsis, various species of Callithamnion, 

 the hairs of Chondria, Polysiphonia, Griffith sia, etc., but death 

 may occur without such extreme physical changes in the proto- 

 plasm, as in the cells of the thallus of Griffithsia. It seems prob- 

 able that the changes characteristic of Bryopsis, Callithamnion 

 and of the hairs of various forms occur where the layer of proto- 

 plasm is very thin and perhaps contains a high percentage of 

 water, while the cells with thicker or a less fluid wall die without 

 exhibiting such extreme physical changes. 



The observations on susceptibility gradients in single cells in 

 Bryopsis, Callithamnion and Griffithsia show very clearly that in 



