284 
DETERMINATION OF MOISTURE IN TWIGS AND FLOWERS. 
RAE Pipe ise ion) 
F |o Ie] S| & | &s 
as eo |?) a7] Ba o of 
Hn D nS 
Date. VARIETY. a | foi esl se] sh 
Ce PACHA Pd Pec ia te ai le 
Ee Phy a he Pee <q 
Apr. 30 |Pear—Bartlett................. 19.78} 6.24 i 2.9 ccm 68.7 
“ AVG aha sh Ade abe 24.2 | 7.55) 8 3.5 | 68.8 1 
4 ASW BOCK Oks. tect as cl iene s 28.03] 9.52)11| 3.9 | 66. ! 67.3: 
si Ss Se a SED) haces fash culate lace 24.78] 7.79) 9} 3.6 | 68.5 ; 
a ‘i Flemish Beauty eniereinee aate 25.07) 6.99) 8] 3.1 at 72 4 
= ee Ue. ean a RMI WEY Be aeaeeity ate 22.54] 6.15) 6) 3. Tay ; 
oy ULC MBOUSSOCK 10h PCA Ee 37. 78/11.11)10) 3.3 ae "1.9 
vy ¥ “OY OF bisector. Soph 39.67)11.2 |10) 3.4 | 71.8 i 
SS AD Dl Care ALCL WEIN ORE dint ibonaia wa stats 24.12) 9.12) 8) 5. | 62.2) 6, 4 
ty fr te aac aimee & delight 23.88] 9.4 | 8) 3.7 | 60.6 ; 
dh Me PL SSUSSOG) sting Sete was oe aia is 32.2 |14.45) 6] 6. | 55.1 51.7 
ACTA A EE, RIE A, See 36.46/15.81| 7| 5.7 | 59.49 2° 
rT «4 Esopus Spitzenburgh. ...[80.88)11.8 |10) 4.4 aaa 62.2: 
"3 et ....(29.384/10.94 10) 4.5 | 62.7 ; 
‘¢ |Mountain Ash—European PE, Shag st 26.65) 8.37) 7 3.7 Batt 65.3: 
vt nha age asec ae 32. 110.98) 9| 4.4 | 65.7 ( 
A more satisfactory series of percentages, in respect to 
accuracy, was obtained from the fruit. The first lot was. 
taken when the fruit was something over one-half grown, 
and the second lot when full size and approaching ripeness. 
Kach pear was quartered, and the quarters distributed, so- 
that even closer duplicates, if possible, were secured than 
with the twigs. This table represents a relationship that. 
corresponds very well to experience regarding the tendency 
of each variety to blight, with the exception that the Buf- 
fum should stand above the Seckel, instead of being the: 
lowest of the series. But this exception need not be made: 
if we consider the progress of the disease in the fruit itself, 
when introduced by artificial inoculation or otherwise. It 
is not known, however, what relation exists between the 
succulency of the fruit and other parts of the tree, and this. 
table, therefore, helps us but little in solving our problem. 
It is interesting to note that in sixteen determinations of 
the fruit of the Buffum pear, made under as closely identi- 
cal conditions as it was possible to devise, a variation of 1.6: 
per cent. occurs, with an extreme variation from the mean. 
of nine-tenths of one per cent. Using these figures and 
applying the formula,! plus or minus 0.6745 into the square: 
root of reren| in which S is the sum of the squares of the dif- 
ferences between each separate observation and the average: 

1Kohlrausch’s Physical Measurements, Eng. tr., 1874, p. 2. 
