CHRONOLOGY DOUGLASS AND ANTEVS 307 



rings, which by their small size denote drought years. The coast 

 redwood is less satisfactory, but is now receiving adequate study. 

 Earlier groups of these important trees proved to be total failures 

 in cross dating, but a personal inspection of the forests last summer 

 showed at once that the lower parts of the trees are so subject to 

 erratic growth from fires and wind strains that cross dating close 

 to the stump, as is usually done, is utterly out of the question. How- 

 ever, it was evident at once that higher portions of the trunk would 

 be less subject to these injuries and we are actually now succeeding 

 in cross dating between the tops of these trees. Yet, even so, they are 

 not so easily dated as the species already mentioned. That could be 

 due in part to their much divided allegiance, namely, to the rains 

 of winter, the fogs of summer, and their almost continuous growing 

 season throughout the year. 



Trees in north Germany cross date very readily; also trees at the 

 Arctic Circle in north Sweden. The pines of central and southern 

 Sweden do not match as well as hoped, but the spruces cross date 

 accurately. It is logical to suppose that trees in southern Europe or 

 northern Africa could be cross dated as in our corresponding lati- 

 tudes here. Thus many points in widely separated parts of the 

 northern and southern hemispheres are almost certain to give results 

 that will supply valuable climatic information. There is no doubt 

 that the interpretation of ring width in different climatic regions 

 needs a vast amount of detailed study, including the establishment of 

 meteorological stations within the forests chiefly concerned. 



While many regions give interlacing cycles apparently related to 

 the 11-year cycle, certain regions give the 11-year variation without 

 complications. The trees of north Europe, especially near the Baltic 

 Sea, give a very perfect example of this since 1830 in curves whose 

 maxima and minima coincide with those of the sun-spot numbers. 



CYCLE ANALYSIS 



Analysis of tree records has been done chiefly by the cyclograph 

 process, which depends upon a pattern called the cyclogram, in which 

 one can see at once not only the length but also the beginning and 

 ending of each cycle, its steadiness or variability, its change of phase, 

 its composition, and to a considerable extent its amplitude. 



The process could be described as an interference between an 

 approximately perfect period engraved on glass in the form of 

 parallel, equally spaced lines and the observed maxima extended into 

 parallel bands by a cylindrical lens. These two systems of parallels 

 are set at an angle of about 12° to each other. In the pattern pro- 

 duced by the transmission of the latter through the former, the ob- 

 102992—32 21 



