ELEVATION IN A BRANCH OF A HORSE-CHESTNUT TREE. 505 
Finally, the simultaneous fall of both leaves and fruit during 
October is accompanied by an ascent of the branch—the 
Autumnal Rise, one may call it—which is both greater and more 
rapid than any other change in elevation, either ascent or 
descent, which takes place in the course of the entire year, the 
branch recovering, during about four weeks (as may be seen in 
the results of observations 9, 10, 20, 21, 81, 32, 33, and 34), the 
elevation it has been more or less rapidly losing during the pre- 
ceding twenty-five weeks or thereabout. Upon this Autumnal 
Rise succeeds the Winter Rest, already spoken of. 
A disturbing cause (alluded to above) is apparent in con- 
nection with the observations numbered 27,25, 29, and 30, made 
on and between July 18th and September 19th 1897, respec- 
tively. During this period, not only did the usual rapid 
autumn descent not take place, but there was first a slight 
fall, then a period of no change, and, finally, a very slight rise; 
all which movements are in strikiug contrast to what took place 
at the sume periods during the two preceding years. This was 
due, perhaps, in part to the very hot dry weather which prevailed 
during June, and led, doubtless, to the foliage being less 
luxuriant (and therefore lighter) than usual ; but it was certainly 
due, in the main, to the great injury done to the foliage and young 
fruit by the astounding hailstorm and ‘ tornado’ which devastated 
the central part of Essex on June 24th. Although my house 
stands on the extreme margin of the storm-area, and although 
the hailstones which fell around it were of very small size com- 
pared with those which fell only a mile away, the injury to growing 
crops and vegetation of all kinds was exceedingly great, and 
about sixty panes of glass were smashed in my house and green- 
house. The leaves of the chestnut-tree in question were very 
severely cut and slashed and pierced, so that their weight upon 
the branch must have been very substantially less than before 
the storm. I have shown on the Chart, by means of a broken 
line, the course the connecting line would probably have taken 
had it not been for this drought and extraordinary hailstorm. 
My observations, as shown on the Chart (Table B., Pl. 27.), seem 
also to bring out another point in connection with the rise aud fall 
of the branch—namely, that (in addition to its periodic seasonal 
rise and fall, as described above) it is also gradually undergoing 
a permanent descent; for both its maximum and minimum 
elevations showed a tendency to lessen in each succeeding 
