NO. I GROWTH LAYERS IN TREE BRANCHES — CLOCK ET AL. 47 



Poorly developed Hghtwood detracts from contact sharpness to an 

 extent nearly equal to its absence. Under high power, 1939 in TTP 

 2i-8-a is a complete ^ growth layer consisting of two to three rows of 

 wide cells and one to four rows of narrow cells. These sizes not 

 only approach each other here and there around the circuit but also 

 approach the sizes of the cells immediately contiguous to them in 

 adjacent growth layers, a transitional feature making it difficult to 

 recognize 1939 at those places except under high power. The next 

 outward growth layer, 1940, is of very irregular thickness and almost 

 wholly thinner than 1939. Under low power this irregularity is seen 

 as a series of short lenses, but under high power it resolves itself into 

 a continuous band of cells around the circuit. The growth layer may 

 narrow to one or two cells in thickness, these being radially narrow and 

 seemingly a part of the densewood of 1939. In places, the inner mar- 

 gin of 1940 is indistinguishable. It is to be noted that the use of high 

 power eliminates the necessity of assuming that the cambium failed 

 to divide throughout an entire season and of concluding that an an- 

 nual increment is totally absent. 



In XSC i-2-b the outer part of 1935 consists of two entire growth 

 layers. Both outer contacts are sharp except on the short radius where 

 the inner growth layer (under low power) looks less sharp than the 

 outer because the Hghtwood of the outer is reduced to a thickness of 

 one cell which is not as thick everywhere as the Hghtwood cells at the 

 start of 1936. The point is worth emphasis : the definiteness of a 

 growth layer may depend more upon the characteristics of the suc- 

 ceeding Hghtwood than it does upon its own characteristics, even 

 though these are definite and sharp. Close study under high power 

 is necessary to reveal true character and definition. Many examples 

 of this have been noted and how many more of the growth layers 

 classed as lenses in our collection actually are entire and complete, only 

 prolonged detailed analysis would reveal. 



One more example, TTP 23-4-c (pi. 17, fig. 2), will be sufficient to 

 illustrate the role played by atypical Hghtwood in marginal definition. 

 The growth layer for 1940 extends four-fifths of the way around the 

 circuit and is made up of Hghtwood and densewood. Over the rest of 

 the circuit, under low power, the growth layer does not exist — hence 

 1940 would be interpreted as a lens. High power reveals, however, 

 that over this portion of the circuit 1940 is represented by at least one 

 row of densewood cells. Furthermore, there is variation between 



* The term complete refers to the radial development of a growth layer and is 

 distinct from the term entire which refers to its areal development around the 

 circuit. 



