3O C. M. CHILD. 



limits a very delicate indicator of physiological condition and partic- 

 ularly of rate of cell respiration. When susceptibility gradients 

 are found to be essentially the same for different agents which act 

 on living protoplasm in ways as different as KCN, HgCL and 

 the vital dyes in the concentrations used, it seems impossible that 

 the differences in susceptibility can be due simply to differences in 

 permeability. In this work on the hydroids the use of a large 

 number of agents for lethal susceptibility was regarded as unneces- 

 sary for several reasons : first, because differential susceptibility 

 to KCN is believed to be a valuable indication of quantitative dif- 

 ferences in physiological condition which in most, if not in all 

 protoplasms, are associated with oxidative processes; second, be- 

 cause the data obtained by different methods, lethal susceptibility, 

 reduction and modification of development, the last to be con- 

 sidered in a later paper, all agree as regards the physiological 

 gradients indicated ; and third, because in the light of earlier work 

 on many different forms, with the methods used here and with 

 other methods, the data on hydroid development appear merely 

 as additional evidence in agreement with that already at hand. 



And finally, the physiological gradients demonstrated experimen- 

 tally correspond to differences in rate of the developmental proc- 

 esses. The high ends of the gradients are the most active 

 regions. Differences in permeability are in all probability con- 

 cerned in the differences in activity, but if this is the case, perme- 

 ability is merely one factor in the complex of metabolic and 

 other factors which constitute physiological condition. 



The basis of the general relation between physiological condi- 

 tion and susceptibility has been discussed elsewhere. At present 

 it need only be pointed out that it does not depend on the particu- 

 lar method of action of a particular agent, but is apparently a 

 relation between a certain degree of disturbance of any kind and 

 the rate of change characteristic of the system disturbed. To a 

 degree of disturbance adequate to bring about disruption or ir- 

 reversible change of the system in course of time, the more active 

 region is more susceptible because its own activity becomes a fac- 

 tor in producing the total or final effect (Giild. '23^). 



As regards the gradients indicated by reduction of KMnO,, it 

 has already been noted that they appear both as gradients in rate 



