December i, i38i.] THE TROPICAL AGRICULTURIST. 
527 
an autumnal leaf in Europe, and fell soon after, about 
May* 3rd, having done its work for the tree, but 
having escaped the ravages of the mycelium Hernileia :— 
o M 3 
fief 
A. — June 20 
B. — Nov. 22 
C— Dec. 20 
D. — July IK 
E. — July 1G 
F. -Jan. 21 
N 
2 \ p 
Date on or 
about which 
the leaf fell. 
Hernileia spots, 
absent or 
present. 
Length of time 
the leaf 
remained 
(approximate). 
None. 
16 weeks. 
IS 
None. 
None. 
17 ," 
t 
18 
None. 
18 „ 
ese examples should be compared with ex- 
amples G, I, and K, in the last table. It is to be 
noted that none of the small spots taken into account 
here were either very vigorous or present for a long 
time before the leaf fell. It is also worth remark that 
20—21 weeks is the longest period I have yet found 
a coffee leaf to persist on the t rees in the open. 
To determine, as 
and other changes 
deuiya, certain tr 
and watched durii 
notes being made 
ditious of the buds 
coloured ribbons hac 
The shortest tin 
developed in my < 
•when the bud tirst 
adult pair of leave; 
period occupied by 
very abnormal exan 
pied iu the prooess, 
appeared free on 0 
until the following 
fnlly formed before 1 
During the very 
deniya from Decen; 
Whole much less ai 
of leaf- buds and le 
April to AuguBt, an 
general experience 
Nuwara Khya. Of 1 
rate of devclopmen 
Stances other than 
which pruning, mai 
and other phonome 
■lower in developin 
temperature prevail 
•warm, it will be e\ 
other processes of 
observations were 1 
and bear little crop, 
taken into account :- 
as possible, the rate of growth 
leaf, buds, and leaves at Pora- 
were selected in August 1S80, 
the following twelve months, 
short intervals as to the con- 
d leaves ou twigs around which 
ien loosely tied. 
in which a pair of leaves was 
eriinents was about four weeks, 
posed, on April 10th, became an 
y May 8 th ; and the lougest 
1 same process was, excluding a 
3 where 17 -i weeks were occu- 
weeks, where a bud which first 
ber 29, did not begin to open 
inuarv, and its leaves were not 
f weather experienced in Pera- 
■ to March, there is on the 
ity displayed in the formation 
s, than during the period from 
this is in accordance with the 
jnt. where winds or a general low 
t hau where the air is quiet and 
ident that elevation affects this as 
growth. The trees ou which my 
iade 811 tier much from leaf-disease, 
circumstances which must also bo 
llniwli A. ( Tree not Pruned. ) 
Aug. 
Bept. 
Oct. 
Uar.li 
April 
Mm 
Juno 
W 
2 An.,-. 20 Sept. 
" Oct. 
I-Yli 
Feb. 
April 
May 
Marin BO 
May 
June 
ill"* §1. 
of SI 
U0 July UiJuly 
0 Oct. 10 7 
28 Nov. 161 \ 
1 Fob. 27 7 
Oi March 2H s 
6 1 April BS I 
'J.linic I., :\\ 
HOJuly 111 \ 
lH.Vng. i| :<) 
weoks. I 10J weoks. 
On this branch, therefore, eight pairs of leaves were 
completely formed during the 12 months (August I, 
1S80, to August 1, 1881), the period of slowest growth 
being September to February ; that of qnicke-t growth, 
April to August. The pair of leaves which developed 
most rapidly was formed in June — July ; that which 
developed most slowly, in January — February : — 
Branch B. (Not Pruned). 
Date on 
which bud 
became 
visible. 
Date about 
which bud 
began to 
open. 
Date when 
leaves had 
separated. 
Nov. 8 
Feb. 25 
April 25 
April :)0 
May 2 
May 30 
Aug. 8 
Time occu- 
pied in 
opening 
(approxim- 
ate). 
(WlJ 
Aug. 6 
Nov. 1 
March 7 
March :i0 
April 20 
April 30 
Oct. 10 
Jan, 20 
Feb. 21 
March — 
April 14 
April 25 
luly G 
Oct. 1G 
March 7 
March :J0 
April 20 
May 1 
July 20 
i weeks. 
4i ' » 
0" 
2* 
4* 
12 weeks. 
17* „ 
7* „ 
6 
Branch C. (Pruned Tree). 
a'S 0 . 1 ^% 0 
P j p p 
Date when 
leaves had 
separated. 
Date on 
leaves had 
attained 
full size. 
&1 s §« 
Jan. 22 Jan 30 Feb. 10 Feb. 27 i weeks. 
Jan. 3l|Feb. lo!Feb. 21 March 14 
Feb. 24|April 2lApril 15fMay 
April 10 — — May 
May IMay 8 May 20 June 20 G weeks. 
May 25|.Iune 15;June 20 July 
E. 
The fact that the spores of Hernileia are convey- 
ed from place to place by wind has been abund- 
antly proved. I select the following as illustrations 
of experiments equally successful in Badulla, Diinbula, 
and other districts besides those named. 
Clean slips of glass measuring 3 in. by 1 in., and 
therefore approximating the area of a coffee leaf of 
small size, wore smeared evenly on one side with a 
thin layer of pure glycerine, and exposed in various 
positions for various lengths of time, in such a man- 
ner that spores borne by the wind to the viscid 
surface would be retained, and afterwards detected 
by microscopic examination. 
The examples chosen for the following table illus- 
trate more or less the following conclusions : that 
(1) the spores are borne by wind in all parts of 
country, (2) the more "rust" present, the larger 
the number of spores in the air. 
To illustrate further, No. 5 may be selected as re- 
markable. In Peradeniya, on July 8th, 1 88 1, the trees 
were much diseased and "rust- patches" showing up 
badly all over, the wind was very high, and blowing 
from the S.W. in heavy gusts. One of the slips o£ 
glass was placed vertically on end in a cleft bimboo 
and arranged so that the viscid surface faced the 
nearest coffee at a distance of 18 feet-. The length of 
the b imboo was a little over 5 feet. After 12 hours' 
exposure the slip was examined, and 117 spores of 
Hernileia found imbedded in the glycerine, together 
with small insects, insects' eggs, spor. s of various, 
fungi, alga), aud lichens, and pieces of sand, &c,, 
blown iu the same manner on to tho trap set 
Another proof of tho same fact is afforded by my 
lindiug numerous spores of ffemiteia entrapped in the 
meshes of »orae canvas exposed amongst the coffee for 
several mouths : there were spores iu all stage* of 
