NO. I GROWTH LAYERS IN TREE BRANCHES — CLOCK ET AL. 273 



from simple to complex, on TTAp 2-3-b (text fig. 38), TTC 33-6-a 

 (text fig. 5), TTC 12-9-b (text fig. 33), XSC 2-2-c (text fig. 40), 

 and TTC 36-7-b (text fig. 25). Clearly enough, the student should 

 be well aware of the disadvantages of a single radius or of a cross 

 section and should buttress his work by a thorough investigation of 

 the growth habits of the trees with which he works. 



Visual problems. — Stained sections under a microscope have many 

 advantages over unstained sections under a hand lens or wide-field 

 binocular, especially for anatomical work. These advantages are well 

 known to botanists. 



Although lumens are visible under a hand lens if the surface has 

 been well prepared, such visibility cannot, of course, be compared with 

 that under a compound microscope. The densewood, its resolution and 

 its outward termination, must be seen with all possible clarity. The 

 margin of any given growth layer appears to increase in sharpness as 

 power af magnification used decreases, unless the cells of the light- 

 wood immediately outside the margin are unduly lignified. In this 

 latter case the true sharpness only comes out under high powers. A 

 hand lens has two advantages: (i) it is easy to carry, and (2) in a 

 way, it summarizes anatomical features by omitting details. These 

 details, however, may carry highly important ecologic information. 



Extremes of multiplicity. — Previous pages have carried information 

 on the multiplicity typical of the accentuated forest-border environ- 

 ment on the southern High Plains. Certain species and certain years 

 show extreme numbers of intra-annuals whereas others show only a 

 few. In some species, it is impossible to find a year lacking multiple 

 growth layers. The reason for such multiplicity has commonly been 

 held to be large and repeated fluctuations of soil moisture. 



The practice called tree-ring analysis, or dendrochronology, has 

 presumed accurate dating to the year centuries in the past by the 

 method of matching thin growth layers in one tree with those in other 

 trees, whether they grew close together or tens of miles distant. If an 

 extremely thin, entire growth layer, or a lens, appears in a certain 

 place on the sequence in one or several trees and does not appear in 

 the other trees of a group, it is designated a "missing" ring and is 

 added to the sequences of the latter trees in order to fill out the sup- 

 posedly correct number of "years." This undoubtedly is what Doug- 

 lass (1936, p. 12) meant when he said, "At the very desert edge the 

 trees may become difficult from absence of rings." Perhaps it is not 

 the lack of rings in some of the trees so much as it is their super- 

 abundance in other trees that creates difficulty and often leads to in- 

 correct dating. Many trees at the edge of the desert possess so many 



