468 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1955 



Since the actual amount of annual radial gi-owth of the stem is a 

 function of the tree's age, species, environment, and other factors, 

 the growth must be expressed in departures from the mean trend, 

 preferably as percentages, in order to derive climatic indices. Obvi- 

 ously the position of the trend line fitted to the data may be raised 

 or lowered if the series is extended; this so-called end effect is usu- 

 ally of little importance in the outer part of the growth curve of a 

 mature tree, where the mean growth rate has usually approached 

 a nearly constant value, but may be critical in the early part of the 

 record, where the rate is usually changing fairly rapidly. Since 

 the age trend in over-age chronology trees is especially shallow except 

 in the earliest decades of growth, these trees can provide indices 

 which represent a good approximation to absolute values. On the 

 other hand, it is evident that any small secular trend in climate 

 would be completely hidden in the fitted trend line and thus would 

 not be determinable even in these trees. 



A typical site on which such trees may be found is illustrated in 

 plate 3. For many miles bordering the upper Colorado River, as in 

 this area just west of Eagle, Colo., stunted Douglas firs and pinyon 

 pines {P. edulis Engelm.) dot the steep slopes and are readily ac- 

 cessible from the highway. Standing dead poles are comjnon. That 

 this site happens to be a gypsum formation may have no great signifi- 

 cance, smce trees of comparable age, sensitivity, and slow growth 

 have been fomid on sandstone or limestone slopes nearby; however, 

 the number of extremely old Douglas firs per unit area is greater 

 here than at any other site in the Rockies thus far sampled by the 

 writer. A closer view of one of the trees in plate 3 is shown in plate 4, 

 left. The dead snag above a small cluster of live limbs is character- 

 istic of many over-age Douglas firs. 



Another excellent source for drought-chronology trees is the Bryce 

 Canyon National Park area. The stunted Douglas fir in plate 4, 

 center, photographed in its dying years, has for some seven centuries 

 been overlooking what is now called Bryce Point. In plate 4, riglit, 

 the roots of this tree, exposed for hundreds of years, show a 2- foot loss 

 of soil during the tree's life. The increment borer, which makes avail- 

 able 15-inch cores, provides a scale. 



On the basis of several hundred sampled trees which were suitable 

 for chronology studies and which exceeded the coimnonly assumed 

 maximum age of about 500 years for Rocky Mountain conifers, cer- 

 tain characteristics of these drought-site trees emerge as probably 

 very general in nature: {a) The absolute maximum ages of Douglas 

 fir, ponderosa pine and pinyon pine in the Rocky Mountains area of the 

 order of 1,000 years, and for P. flexilis (limber pine) in excess of 



