SECT. 2] AND WEIGHT 459 



are much the same when the embryo has o somites as when 

 it has 20, but, in view of the increase in total length during this 

 time, it will be seen that the maximum variation is much less 

 important as time goes on. "The relative differences", said Fischel, 

 "are certainly less in the later stages than in the earlier ones." 

 Fischel's measurements of the lengths of the parts all showed the 

 same relation as regards variability, but, though the length of 

 the body is increasing regularly all through this period, the length 

 of the part between the anterior end and the ist somite remains 

 practically stationary, as does the length of the part between the 

 last somite and the posterior end of the embryo. In other words, 

 the increase in length is entirely due to growth of the middle region 

 in which the somites are being produced.* The size of the individual 

 variations can be large; thus an embryo may be more than 50 per 

 cent, longer than another one of the same stage. Fischel's examina- 

 tion of the work of other authors, such as that of Bonnet on the sheep 

 and of Keibel on the pig, induced him to suppose that very similar 

 effects were seen in mammalian embryos. Philiptschenko carried the 

 question into insect development by investigating an apterygote, 

 Isotoma cinera, and found that the older stages showed the greater 

 variability. On the other hand, Zuitin, who has studied the develop- 

 ment of Dixippus morosus from this point of view, found that, just as 

 in the birds and mammals, the earlier stages were the ones which 

 showed most variations from the mean. Schmalhausen has also con- 

 sidered the question on the basis of the figures he obtained in his 

 studies on the growth of the chick embryo, already referred to. He 

 points out that the early stages in embryonic development are those 

 when the anlages are being formed. At that time, every few hours, 

 as it were, are marked by the start off of one or more parts or organs 

 on the long declining hyperbola which represents their instantaneous 

 growth-rate. Thus a cross-section through an embryo in those early 

 periods would demonstrate, if some method of the future made it 

 possible, a series of growth-rates, some low, appertaining to the more 

 senior organs, some high, appertaining to the more junior ones. 

 Moreover, the mass of each anlage is different for each organ, so 

 that extremely complicated effects will be observed if the weights of 

 the organs are described in percentage of the total body-weight. Thus 



* As Levi has shown, embryo size in the somite stage is very similar no matter what 

 the size of the adult bird, but soon the larger animals grow longer. Also the size of the 

 first somites of all sauropsida is between 5000 and 8000 /x^ surface. 



