NO. I GROWTH LAYERS IN TREE BRANCHES — CLOCK ET AL. 1 53 



Tree TTP 23 shows rather remarkable consistency of growth- 

 layer patterns among the sections of each branch. In TTP 23-1, 2, 3, 

 and 4, the 1940 increment appears at first sight to be a problem 

 growth layer of great uncertainty. TTP 23-4 (table 88) serves as 

 an example of all these branches. In sections a, at 34.5 cm., 1940 has 

 been reduced nearly to a minimum. It appears as divided densewood 

 of 1939 at one or tw^o places, dependent upon the particular section 

 examined. At these places, 1940 consists of one to two rows of large 

 cells and one to two rows of narrow cells. Elsewhere no lightwood is 

 present, only densewood. The reduction has gone so far at some 



Table 

 34.5 cm. 



1938 I msee 



1939 I msce 



1940 I ee 



1941 inc L 



points that the densewood of 1939 plus all of 1940 consists of two 

 narrow cells, one for 1939 and one for all of 1940. Magnifications 

 of 20X or less do not reveal the presence of 1940 — analysis at such 

 powers would no doubt consider 1940 absent. In sections h, at 21.3 

 cm., the 1940 increment is clearly present around the entire circuit. 

 For at least tv/o-thirds of the distance, it consists of one wide cell 

 and two narrow cells; for the other third, it is up to five cells thick. 

 Sections c, at 13.8 cm., show 1940, under low power, as a lens cover- 

 ing four-fifths or more of the circuit. High power reveals it as an 

 entire growth layer, densewood being continuous. In addition to the 

 gradual radial reduction of the growth layer near the cusps of the 

 "low-power" lens, there are other irregularities of length in adjacent 

 radial columns of cells; for instance, one wide and one narrow cell 

 make up 1940, or three narrow cells comprise densewood of 1939 

 plus all of 1940, or one densewood cell makes up all of 1940. 



The increment of 1940 is best developed in sections b and least de- 

 veloped in sections a. If the several wide cells are absent from sec- 

 tions a, 1940 would be represented by densewood only and therefore 

 would be indistinguishable from the densewood of 1939. The light- 

 wood of 1941 is so large-celled as to give a striking contrast with all 

 of 1940. If these growth layers were not dated accurately and 

 studied closely under high magnification, some workers might be 

 loath to call 1940 an annual. This applies here as well as to many 



