Mar. 1892.] THE BOTANICAL SOCIETY OF EDINBUROH. 8^1 



in a regular proportion, often indeed very irregularly. But 

 to this rule certain species ofter a remarkable exception. 

 In the genus Quercus, and especially in Q. Cerris, a decided 

 tendency to halt in mid-season was shown, and this was still 

 more marked in the genus Abies, and particularly in Ah. 

 Loiviana and grandis. My weekly observations show that 

 in the latter there was not- merely a retardation of growth in 

 mid-season, but an actual stoppage for at least two weeks in 

 two successive years. 



From this and other facts which do not come within the 

 scope of this paper it appears that the progressive rate of 

 increase in girth, in some species at least, follows laws which 

 to a great extent are independent of temperature or other 

 meteorological causes. 



Distribution of Girth- Increase over the Growing Months.— z- 

 The maximum of girth-increase falls on very different 

 months in the difierent species, but is less widely distri- 

 buted in the Deciduous than in the Evergreen class. In 

 the former it falls almost always on June or July, the only 

 exceptions in my old and young trees being Liriodendron 

 tuliinferum, where it falls on August, no doubt normally, and 

 Carpinus hctulus, No. 86, and Fopulns fastigiata, No. 76, in 

 which it also falls on August, but probably abnormally ; in 

 several other instances, however, the proportions for July and 

 August are equal, or very nearly so. , But in the Conifera?, 

 although June and July are also highly favoured, a con- 

 siderable number of the maxima fall very regularly and 

 decidedly on May, and in Ah. Donglasii, No. 99, and grandis. 

 No. 91, August takes the first place. 



Generally the girth-increase is more evenly distributed 

 over the season in the Coniferae than in the Deciduous 

 species. 



As a rule, the Deciduous species get over the main part 

 of their growth earlier than the ConifercT, but there are 

 remarkable exceptions in both classes. For example, nearly 

 four-fifths of the increase in Araucaria imhricata took place 

 in the first half of the season, and nine-tenths of the increase 

 in TAriodcndTon tvJipifcricm in the last half. 



Connection of Girth- Increase with Development of the 

 Zm/— The tendency in the species to an early or late 

 preponderance of girth -in crease is not necessarily dependent 



