THE MOSAIC STAGE OF DIFFERENTIATION 221 



a certain point without preventing the construction. It may have 

 too much wood and not enough glass, or too much glass and not 

 enough wood, but, provided that the disproportion does not exceed 

 a certain degree, it will still be a conservatory. But if the amount 

 of glass be too great for the wood, the construction is mechanically 

 impossible. The yolk and fat in the frog's egg may fancifully be 

 compared with the wood and glass in the conservatory. 



Other evidence of a similar nature is provided by centrifuge 

 experiments on the eggs of echinoderms, in which centrifugation 

 has been continued until the tgg has separated into two or even 

 four (unequal) portions along the direction of centrifugal force 

 (which of course may bear any relation to the original polarity of 

 the egg). The fragm.ents differ considerably in colour and the type 

 of their contained granules. We may call these halves A and B, and 

 the quarters A^, A^, B^, B2, in order from centripetal to centrifugal 

 region. In Sphaer echinus gramdatus, a centrifugal half {B) or either 

 of the two centrifugal quarters {B^ or ^2) is capable of producing 

 plutei. A centripetal half {A) on the other hand never goes further 

 than the blastula, and the same is true for the most centripetal 

 quarter (^1). The other quarter {A^, however, may in some cases 

 produce a pluteus. We may provisionally assume that the fragments 

 incapable of pluteus-formation contain an excess or defect of certain 

 raw materials, as in the frog experiments described above. Pre- 

 sumably the excess substance responsible for failure to develop in 

 the A halves was all contained in the A^ quarters, thus permitting 

 the A2 quarters to develop. Results similar in principle have been 

 obtained for several other genera : in one case ( Tripneiistes esculentes) 

 the conditions are reversed, the A (centripetal) pieces being capable 

 of fuller development.^ 



§5 



The determination and localisation of organ-rudiments is revealed 

 sooner or later by the presence of chemo-differentiated material or 

 morphogenetic substances in certain places which constitute what 

 may be called fields, or areas of differentiation of organs. Within 

 the fields the presumptive rudiments become determined by pro- 

 gressive chemo- differentiation. As an illustration of this important 



1 Harvey, 1933. 



