146 A Provisional Hypothesis of Pangenesis. { March, 
selves, we should expect them to conform to the same laws. Most 
of the cells of the body are at any given time very perfectly 
adapted to the conditions under which they are placed, that is, 
such an adjustment has been brought about during the process 
of evolution of the organism, as to place each cell under such 
relations to its environment as are most favorable to the perform- 
ance of its function in the body. This state of things will last 
until some unfavorable change takes place in the environment, 
either external or internal to the body. The adjustment be- 
tween the cell thus affected and its conditions will of course be 
disturbed by the change, and if this change is great enough to 
check the performance of its normal functions, but not sufti- 
ciently great to destroy life, the cell will, after the analogy of 
other organisms, give birth to gemmules. As these gemmules 
when transmitted to the next generation are supposed to give 
rise to variations, we have a simple and consistent explanation of 
what is without doubt the greatest difficulty of the theory of 
natural selection: how, among the countless numbers of possible 
variations, a given cell ever happens to vary at the time change 
is needed. This explanation is all the more satisfactory since it 
simply embraces the unicellular organisms which compose the 
body under laws which are well established as applied to inde- 
pendent organisms. We can also understand why variations do 
not usually make their appearance in the individuals upon which 
the new conditions are first brought to bear, but in succeeding 
generations; for the new conditions do not result in direct varia- 
tion, but in the production of gemmules which are transmitted 
to the next generation. It may perhaps be asked why a cell 
produced from a gemmule should be more variable than one pro- 
duced by division. A cell formed by division commences its 
existence as a fully formed cell, but a gemmule has the absorp- 
tion of food and the building of a body still before it, and it will 
therefore be more susceptible to external conditions, just as a 
house in process of construction is more easily altered than one 
which is finished. 
If our assumption that newly acquired characteristics are 
transmitted by the male and those of long standing by the fe- 
male is correct, the phenomena of crossing should furnish us with 
a test of the hypothesis. According to the theory of evolution, 
animals of allied species and varieties are the descendants Of & 
common ancestor, and those characteristics which they have 1 
common are due to this community of descent and are of long 
