322 Philippine Journal of Science 191/ 
Table III. — Temperature for periods of four weeks in forest at the top 
of Mount Bnnahao, Luzon, P. I. Altitude, about 2,100 meters. 
Four weeks ending— 
Maxi- 
mum. 
Mini- 
murp. 
Mean. 
Average daily— 
Maxi- 
mum. 
Mini- 
mum. 
Dec. 1,1916—- 
17.7 
10.6 
14.9 
15.9 
13.3 
Dec. 29,1915 
17.1 
10.0 
13.8 
14.7 
13.1 
Jan. 26,1916 
16.5 
8.3 
13.4 
14.6 
12.0 
Feb. 23.1916.. 
15.8 
7.7 
13.2 
14.2 
12.0 
Mar. 22.1916 
17.8 
5.0 
13.5 
15.0 
12.2 
Apr. 19,1916— — - 
17.1 
10.3 
13.6 
14.6 
12.4 
May 17,1916 - 
19.2 
11. i 
15.0 
16.2 
13.8 
June 14,1916 - - 
18.9 
14.3 
15.1 
17.6 
15.2 
July 12,1913 - 
22.7 
12.5 
15.7 
17.4 
14.8 
Aug. 9,1916 — - 
23.6 
9.2 
16.2 
16.7 
14.1 
Sept. 6,1916 - 
19.2 
12.2 
14.9 
16.1 
14.6 
Oct. 4,1916.... 
17.1 
12.2 
15.8 
15.6 
14.2 
Nov. 1,1916 
17.1 
14.5 
15.6 
16.9 
14.7 
14.6 
16.7 
13.6 

Since no annual growth rings have been observed in the wood 
of Podocarpus inibncatus, the rate of diameter growth can only 
be determined by making periodical measurements of the girth 
of the same trunks. On May 29, 1914, the girths of a number 
of trees of Podocarpus imbricatus were measured with a steel 
tape at breast height, or 1.5 meters above the ground ; and that 
the trees might be again measured at exactly the same point the 
place of measurement was, in each case, indicated by a small 
nail. A year later the trees were again measured and the dif- 
ferences between the two measurements taken to be the rates 
of growth in girth for the year in question. As the climatic 
condition in Luzon for this year was approximately average, it 
is probable that the rates of growth obtained for Podocarpus 
are approximately average rates of growth for this species in 
this locality. The growth figures have been converted into the 
more usual form of rates of diameter growth. The results -are 
presented in Table IV, in which the trees are classified ac- 
cording to diameter classes of 10 centimeters. In order to 
approximate the total ages of trees of different sizes, the 
average annual rate of growth of each 10-centimeter class 
was divided into 10 centimeters and the quotients assumed to 
be the number of years necessary for an average tree to pass 
through the 10-centimeter diameter classes. By summing up 
these quotients we can obtain a figure which represents the age 
of an average tree in any 10-centimeter diameter class; that is, 
the quotient obtained by dividing the annual growth of the 
