234 
R. Ruggt es Gates. 
the development of the pygostyle, but something must in turn 
determine the development of these. Morse concluded from his 
experiments that it is possible to suppress metamorphosis, but 
that it cannot be induced in stages far removed from those in 
which it would ..normally occur. A certain cycle of events determined 
by heredity is necessary before any stimulating agent will cause 
metamorphosis. 
Although we have spoken of recapitulatory characters having 
been at one time terminal stages in the life-cycle, yet it appears that 
they have not always been added terminally but have often been 
intercalated at a subterminal or earlier stage. Thus in plants the 
sporophyte ends inevitably with sporogenesis, so that the lengthening 
of the sporophyte generation must always have taken place through 
the addition of subterminal stages, involving the delay of 
sporogenesis. 1 In animals, too, spermatogenesis or oogenesis is 
usually delayed until the ontogeny is complete, although the 
gi eater freedom of the germ cells in their relation to the soma 
makes possible the occasional occurence of paedogenesis or reproduc¬ 
tion by larvae. 
In concluding this chapter it may be pointed out that the 
phenomena of recapitulation furnish another limitation of the cell 
theory, recapitulatory characters being organismal, and embryonic 
recapitulation apparently involving the inheritance of functional 
modifications. 
1 This is a little different from Bower’s well-known view of the sterilization 
of potentially sporogenous tissue. 
