58 HOW ANIMALS DEVELOP 



the presumptive areas. It may be possible to work 

 out such a map in the near future by growing the 

 embryos in cultures Hke the chicken embryos. It is 

 already quite possible to keep rabbit embryos alive 

 for a few days, so that they develop fairly normally, 

 and it' should not be very difficult to improve this 

 technique slightly. But the suggestion that man will 

 soon be born out of bottles is very optimistic (or 

 pessimistic, according to the point of view). The 

 technique of cultivating embryos will probably 

 remain a useful scientific method for a long time, 

 but it is difficult to imagine it being used practically. 

 The mammahan embryo, including the human 

 embryo, therefore goes through a primitive streak 

 stage very like a chicken. Although we do not yet 

 know much about the details of what is happening, 

 there are various reasons for believing that the 

 movements which take place are similar in both 

 sorts of animals. We can draw this conclusion from 

 some of the kinds of abnormal embryos which are 

 found. Gastrulation is a process involving such 

 complicated movements that it is quite easily upset. 

 For instance, the material which moves forward to 

 build up the primitive streak may for some reason 

 go awry and split into two separate streams, so that 

 it makes two front ends of the streak, and the 

 embryo develops with two heads. Or a similar 

 spHtting may happen later, when the presumptive 

 neural plate and body-segments are being pulled 

 backwards into their final positions, and we then 

 get two posterior ends in the embryo, which is split 



