ANIMAL REPRODUCTIONS. 47 
The members in miniature, which, though 
themfelves mutilated, produce a fimilar, but 
fmaller miniature, and the reproductions of this, 
when mutilated, afford a powerful fupport of the 
hypothefis of involution. However, [ will not 
affirm that the reparatory germs are involved in 
each other ; the expreflion would be incorrect ; 
but it appears that the expanding germ includes 
all the parts fit to reproduce a member, and, 
with thefe parts, germs united to them, which 
alfo grow, and that by their connection, and are 
deftined to replace the loft members. In fhort, 
different generations of germs muft not be con- 
ceived as boxes cafed in each other. The germ 
of the fecond generation conftitutes part of the 
germ of the firft, juft as a feed growing in a 
plant, or an egg in an animal, forms part of that 
plant or animal. The germ of the third gene- 
ration conftitutes part of the fecond, and fo on 
with the reft.  Vhus all the generations includ- 
ed in the firft germ are as many decreafing parts 
of it, and it is itfelf a conflituent part of the ori- 
ginal member. I need not return to the objec- 
tions again{t involution; they are anfwered elfe- 
where ; and none fimilar will occur to a philofo- 
pher who knows the difference Hettieas hypo- 
thefis and difcovery. 
You. I. Dd... ., MEMOIR 
