EVOLUTIONARY STATUS OF POLYCOTYLEDONY 
III 
with certainty; lot B, embryos like figures 4, 5, 6, and 10, in which there 
were evidences of either fusing or splitting cotyledons in the primordia 
stage; lot C, embryos of the same age as B, but showing no evidence of 
fusion by the grouping of their primordia; lot D, the oldest embryos, all 
older than B or C, in which, although the cotyledons were not fully devel- 
oped, they were undoubtedly beyond- the critical stages when fusions were 
found to occur. Figures 7, 8, 21, and 22 are from lot C and figures 9 and 
10 from lot D. 
The table below summarizes the result of this study. In lot B, all the 
primordial lobes were counted, even though in some instances they had 
practically fused with a neighbor, but broad cotyledon primordia which 
had no double tip were regarded as single primordia. 
Table I. The distribution of the cotyledons and cotyledon primordia in Picea mariana. 
Cotyledons or Primordia, Number 
Lot B, 23 Embryos, 
Frequency 
Lot C, 83 Embryos, 
Frequency 
Lot 0,63 Embryos, 
Frequency 
3 
0 
9 
5 
4 
6 
53 
42 
5 
18 
20 
15 
6 
4 
I 
Total primordia or cotyledons 
138 
345 
264 
4.93 ± -39 
4.16 ± .41 
4.17 =t .39 
.59 
.61 
.58 
Since lot D represents a late stage, and lots B and C represent an early 
stage, the averages of the number of cotyledons show that there is a reduc- 
tion in the cotyledon number in the course of their development. Of 
course, this indicates that the double cotyledon primordia of the embryos 
in lot B are fusing. 
It will be seen that if these were not fusing cotyledons, but in the course 
of separation, then the oldest group, lot D, might be expected to have an 
average of about 4.93 cotyledons, the average of lot B, or even more. The 
assumption that the embryos which were selected for lot C were not under- 
gomg a fusion of their cotyledons was probably correct, since their average 
number of primordia, 4.16, agrees very closely with 4.17, the average num- 
ber of cotyledons in lot D. However, lot B had an average of 0.76 cotyle- 
dons too many per embryo to agree with the number found in lot D. 
A more careful analysis of the embryos of lot B showed that they could 
be classified in five categories, according to the number of primordia and 
the manner in which these were appearing to unite in forming cotyledons. 
On the twenty-eight embryos of the lot, forty-five double primordia were 
found, which ranged from such cases as were noticeable only when the em- 
bryos were held in a certain position, to others in which the fusion was 
nearly complete and the upper edge of the young cotyledon was only slightly 
retuse. The diagram of figure 6a illustrates the characters of the five gen- 
