80 NATURE AND LIFE. 
proved. It is very plainly seen that cells which have be- 
come individualized by splitting up, are the seat of a di- 
vision which gives rise to other cells; but that only occurs 
when the mother-cells have gained or exceeded their com- 
plete development and their regular dimensions. Now 
this fact, which has been taken as the starting-point of the 
cellular theory, is a mere phenomenon of evolution, and 
not a fact of production. The inventors of that theory, 
for want of observing closely and continuously enough, 
have quite as widely misunderstood what takes place when 
we see certain anatomical elements succeed to others of a 
different kind, as in the case of the liquefaction of one set 
of elements, and then the formation of a blastema in which 
the second set is produced. This is a real genesis by sub- 
stitution, as Robin calls it, and not a direct emission, a 
proliferation, as is taught in the schools across the Rhine. 
In this case there are several phases which have escaped 
the observation of the too systematic doctors of Berlin 
and Wurzburg, but which French savants have settled in 
a way not to be gainsaid, not being blinded, like the for- 
mer, by a preconceived idea. What these same Germans 
call endogenous generation, that is, generation within a 
cell, is quite as much an exceptional mode of production of 
anatomical elements, but one that in no way contradicts 
those we have enumerated, and in no way avails to prop up 
Schwann’s theory. The cellular theory is a doctrine as de- 
lusive as it is convenient and attractive. It is one of the 
numerous mistakes introduced into German science by 
that philosophy of Nature so highly relished hy the con- 
temporaries of Schelling and Oken, and of which the traces 
are to be found in the works of many eminent savants of 
Germany even at this day. Flattering as it is to that in- 
clination by which we are led to the desire to confound 
things the most disparate in one chimerical unity, it is not 
surprising that such a philosophy should so long have im- 
