1904] COPELAND—VARIATION OF CALIFORNIA PLANTS 421 
ties, and even mental eccentricities, are to be inherited. On the other 
hand, mutations are not always inherited, as DE Vrigs’s observations 
on Oenothera show; and if they were, there could be no mutations. 
Variations certainly differ in the reliability with which they are 
inherited; but mutations, if there were such, would not be distin- 
guishable from other variations in this respect, unless sometimes in 
degree. 
Many authors have sought to distinguish single variations or dis- 
Continuous variations from the continuous individual variations by the 
*xtent of the deviation from the parental type. DE VRIES does not 
lay himself liable on this point, saying explicitly that they are not dis- 
tnguishable in this way (p. 41): “Die Betrachtung mancher sm- 
gle variations hat die Einsicht eingebiirgert, dass die Mutationen 
Jedesmal bedeutende Veranderungen sein miissen, namentlich, dass 
‘e gtosser sein sollten als die Variationen. Solches ist aber durchaus 
mich der Fall, und anscheinend sind wenigstens zahlreiche Muta- 
ionen kleiner als die Unterscheide zwischen extremen Varianten.” 
If mutations cannot be recognized by their range of deviation, 
nor by their being inherited, from other variations which may chance 
lo be ‘nusually wide and to be hereditary, there is no test by which 
they can be recognized. If a practical definition of a mutation had 
ever been framed, it could not have escaped DE VRIES; and if his idea 
Could be formulated so that it would represent a distinct phenomenon 
ecognizable as such in nature, he would certainly have given it that 
om. I agree heartily with its friends in welcoming DE VRIES’S work 
4 og Most valuable empirical contribution to our knowledge of the 
eae of novel forms of organisms since the Origin of Species; and 
_ Vates’s method—the analysis of the composite character ee 
*S Into its elements, and the study of the origin (and change) © 
& ie far more rational and promising than the study of the 
SPecies,” as we recognize them, as a whole. But I regard sn eT 
Ons as 8enerically different from ordinary variations, and his specific 
Saacterstics as distinct and clear-cut in their existence and abrupt 
“lag as undefined and not scientifically definable, because not 
“senting distinct natural phenomena. 
ts recognition that the discontinuity 
Rs does not necessarily distinguish them 
of “discontinuous ” 
from “continuous” 
