304 THOMAS C. CHAMBERLIN 
Now it will be seen that this second chain of derivatives is very 
different from the preceding chain and that the two are mutually 
exclusive. The links of the two cannot be mixed without the loss 
of all logical force. Ti mixed, the terms of the problem become a 
hybrid of incompatibles; such a chain does not exist in nature; 
it is not a real problem at all; it is merely a supposititious combina- 
tion of incongruities. 
The concept of isostasy gives rise to one set of derivatives, if 
based on the hypothesis of a crust floating on a liquid substratum 
inclosing a centrosphere of concentric homogeneous layers; and 
to quite a different set of derivatives on the hypothesis of a solid 
elastic earth whose internal material is heterogeneous and has 
suffered internal distortion. 
The problem of the saltness of the sea has one set of subconcepts 
if the hydrosphere, at the outset, was as great as now or even 
greater, and quite another set if the hydrosphere started from a 
minimum and has grown steadily ever since and is growing still. 
To mix these throws the whole effort out of court. And so of not 
a few other earth problems of the more complex order. 
In view of the difficulties of meeting the imperative require- 
ments of consistency in working out such complex problems as the 
inner diastrophism of the earth, it is hoped that the following effort 
to reduce to brief convenient form the essential concepts already 
reached in the study of planetesimal accretion will be found service- 
able. They are not a formal summary of the preceding articles 
nor drawn exclusively from them; some of them even have no 
dependence on the planetesimal hypothesis; they are merely 
found to be tributary to a satisfactory concept of megadiastrophism 
under the conditions of accretion. To make the statements brief 
and convenient, qualifications have been largely neglected and 
some statements may seem somewhat too baldly affirmative, 
but recurrence to the fuller discussions will, it is hoped, show that 
reasonable recognition has been made of legitimate grounds of 
doubt and needs of qualification. It is quite impossible here to 
accredit these propositions to those who have done most to develop 
them; they are merely assembled as propositions tentatively 
accepted as groundwork for further study. ~ 
