NO. I GROWTH LAYERS IN TREE BRANCHES — CLOCK ET AL. 45 



1945, give an excellent example of a series of short lenses of added 

 growth by large-lumened cells around the circuit which represent 

 patches of cells in three dimensions. In TTP 20-23-a (cut November 

 17, 1944), the densewood of the outer growth layer is followed out- 

 ward by two to three rows of large, heavily lignified cells making the 

 outer margin indefinite under the cambium and suggesting that 

 another growth layer was in process of formation when growth ceased. 

 This specimen, furthermore, indicates that growth does not have to 

 cease at an accepted point by an accepted group of features such as 

 narrow, thick-walled, heavily lignified cells. The following question 

 might well be asked : If TTP 20-23 had not been cut off in 1944, but 

 say two years later, would it be possible to detect the exact outer mar- 

 gin of 1944 in view of its indefiniteness ? In the case of the postsea- 

 sonal growth on TTC 12-14, what would a further year of growth 

 have done to the margin of 1945 ? 



The above examples illustrate three types of confusion in the de- 

 lineation and determination of precise outer margins whether they are 

 annual or not. Had the branches been sampled a year or so later, the 

 postseasonal growth would have given either one of two, or both, re- 

 sults : an indefinite outer margin to the previous outer growth layer, 

 or the inclusion of the postseasonal growth in the growth of the fol- 

 lowing year. 



The above considerations must be evaluated in any study necessitat- 

 ing the identification of annual increments without the help of absolute 

 dating methods. The outer contact of an annual increment may be in- 

 definite to diffuse anywhere from zero up to 100 percent around the 

 circuit. If the outer contact is identified as being at the base of the 

 postseasonal growth, then that added growth would be taken as part 

 of the next year's increment and would distort any interpretations 

 based upon measured thicknesses of growth layers. 



Marginal definition is also impaired by atypical lightwood. Among 

 the angiosperms (i.e., TTL 5-1-a) a unique situation exists in that one 

 growth layer is composed predominately of vessels and the following 

 growth layer predominately of tracheids (see pis. 3, fig. i ; 36, fig. 2). 

 The growth layers in the sections of TTL 5-1-a were dated with 

 exactness by tip-growth measurements. The vessels of 1943 and the 

 tracheids of 1944, taken together, appear to be one fully developed 

 growth layer, especially because of the indefinite, not to say obscure, 

 boundary between them. Yet the contact is there, the measurements 

 of tip growth demand the presence of two discrete growth layers, and 

 sections 5-1-b, 20 cm. out from 5-1-a, show a more normal develop- 

 ment of 1943 and 1944. 



