DEDIFFERENTIATION IN PEROPHORA 673 
A chemical analogy, for which I am indebted to Mr. H. R. 
Raikes, of Exeter College, Oxford, may help illuminate the 
point. If one equivalent each of hydrochloric acid, boric acid, 
and ammonia are mixed, a negligible amount of boric acid will 
react with the ammonia owing te its small degree of dissocia- 
tion. We may say that the hydrochloric acid is completely 
‘dominant ’ in the system, owing to a greater speed of reaction. 
If, however, the mixture is heated, the more volatile hydrochloric 
acid will be driven off, and the less volatile borie acid left to 
react with the ammonia. This we may call ‘ differential sus- 
ceptibility’ (to rise of temperature) involving ‘ differential 
inhibition’ of one portion of the system, and consequent 
‘reversal of dominance’. If the mixture were contained in 
a very large closed space, cooling after heating would restore 
the original ‘dominance’ of the hydrochloric acid, giving 
a parallel to reversible dedifferentiation. 
The emergence of the cells from the tissues in dedifferentia- 
tion is a phenomenon which deserves further study. Though 
probably by no means universal it is doubtless commoner 
than is generally assumed. It occurs not only in-Perophora 
but also in Hydroids, in Turbellarians, and in Echinoderm 
larvae, and in many cases of actual poisoning, e.g. by mercury 
salts (Child 1917, Huxley 1921b) and other agencies (Gray 1920). 
Once the cells start to emerge they may collect close to their 
place of origin, or if space and means of transport are available, 
be removed to regions at a distance. When the stolon portion 
is large in a Perophora stolon-zooid system, and the heart is 
beating normally, the latter is the case; it is also the case in 
Hydroids when the coenosarecal portion is large in comparison 
with the hydranth. The difference between the two possi- 
bilities appears to be similar to that between a reversible 
chemical reaction when the end-products are not removed, 
and the same when they are removed. Why, in the first case, 
the tissues should not simply resolve themselves into their 
constituent cells in situ is difficult to see, but the fact 
remains that they do not (e.g. Clavellina ; Perophora with 
very small stolon attached, or with circulation stopped by KCl). 
