252 
G. E, Coghill 
with Type 11 . But when the period of asymmetry and the follow- | 
ing interval is taken into account it is clear that the specimens of i 
Type 111 were a much longer time in passing through the periods ! 
represented by Columns C, D, E, F and G than were the specimens 
of Type I in passing through the period of Column G alone. This 
would seem to indicate that the condition of asymmetry is due to a 
precocious development of one side of the neuro-muscular system I 
rather than to a retarded development of the other side. At any i 
rate the sum of the averages in Columns C and D for Type III is | 
greater than the average in Column B for Type II. It would | 
seem altogether improbablej therefore, that a period of asymmetry | 
like that of Type III has been passed over unobserved in Type II. 
While I do not place any great dependence upon this comparison n 
of the averages in the table, I believe they do tend to show that the 
difference between the different types of reaction as observed in these I 
and numerous other embryos is not based upon relative age but upon i 
the relative development, and probably the variable physiological j 
condition, of the various constituent elements of the neuro-muscular I 
system. When a period of asymmetry occurs, it appears before the i 
period of irregularity or regularity, and never follows either of the | 
latter, excepting in rare cases when one or two movements right at 
the beginning of the experiment are at variance with the asym- j 
metry (figs. 3, 4, 5, 6). The asymmetry clearly influences the ! 
irregular reaction in some cases so that the movements toward the ji 
side touched appear to be determined by a partial persistence ij 
asymmetry (fig. 5). But this is not always the case. The period 
of regularity persists, ordinarily, till near the time of swimming, j 
The actual length of the period varies greatly in different speci- j 
mens, but a comparative study of numerous specimens convince |i 
me that the regularity in response is purest for a period of about |t 
48 hours. 1| 
The structural basis for a regular asymmetry in response must be ]) 
in the ascendency of the effector system of one side over that of the |] 
other, rather than in structural difference in the receptor systems |i 
of the two sides. Two facts particularly support this interpreta- 
tion: (i) All spontaneous movements (somatic) that have been 
observed in embryos which conform to a given asymmetry are in 
accordance with the asymmetry in each case, toward the right in- 
dextrally asymmetrical specimens and towards the left in sinistrally 
asymmetrical specimens. (2) In any given asymmetrical embryo 
