148 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. I40 



the branchlet decreased in each successive tip flush. (3) Or, it might 

 be that the three tip flushes were accompanied by diameter flushes 

 whose outer margins, save that of 1944, cannot be identified. In any 

 event, here in two adjacent branchlets there is a situation not easily 

 explained by the influence of external growth factors, unless it be 

 that the rigorous and variable conditions of the extreme lower forest 

 border promote variable anatomical response throughout the plant 

 body. 



Table 79.—TTP 20-32 



i2.g cm. 4.1 cm. TF 



1943 I see I 



1944 I see I see i 



I dee 



In TTP 20-32 (table 79), the diffuse growth layer does not have 

 a tip flush counterpart. The growth layer becomes so diffuse and so 

 weak out at 4.1 cm. that it would be wholly ignored had it not been 

 seen at 12.9 cm. In this case, as in so many others, the intra-annual 

 becomes more definite down the branch. 



TTP 20-33 (table 38) was set out previously on page 109. Two tip 

 flushes were formed in 1943, and the sections at 11 cm. came from 

 the first flush, whereas those at 6.5 cm. came from the second. The 

 diffuse growth layer is really an arc whose cusps very nearly unite. 

 Because the arc at 6.5 cm. is very probably the same one present at 

 II cm., it cannot represent the first tip flush of 1943. The increment 

 for 1944 in TTP 20-33 is represented by one tip flush. The divided 

 densewood at 11 cm. extends farther around the circuit at 6.5 cm., 

 but whether it becomes the arc at 2.7 cm. is unknown. 



Table 



1943 



TF 



I 



1944 I see I see I 



I psce 

 I psL 



In January 1943 twin buds were observed on TTP 20-35, ^"^ 

 thereafter the resulting twin branches were measured. The growth 

 layers exposed when the two branches were sectioned (table 80) were 

 so completely identical that the above table includes both branches. 

 In the two outer sections (one at 1.3 and the other at 3.6 cm.). 



