E2 PROF. JULIUS MACLEOD ON TEN 
TABLE V. 
Mnium punctatum, Hedw. : distribution of the leaves of 11 fertile stems 
among 10 intervals. 
Intervals : T 9. nA 4. b. 6. y? 8. 0,7 IQ 
NAI... DEI: 4.1 4 3» 3 
SN ee Eg 04 PEEL Y 
LESS E T$ à 13204 151 0 d» a 
o E Ec gu. Bop 31.1 4 1. 3 
LB v EUST. £1 NR 9.9 cw 
£c cup Ea i1 9 14 1292 T 154 
UNS EST. I i fu 359 
Sos DICE E Tj ]1 1 LI ub d 
e a. ETC. 1 Pee a 
x oto r j 1 | d | a 
abs. E X 1 ‘een rey 
Me Tett sich | B xu 1 mH 8 19 1 
ANLOP VG. ce 
Dividing in each interval the total length of all the leaves by the number 
of leaves, we find the mean length in each interval: this gives the mean 
interval curve of the whole group. See Table VI. 
TABLE VI. 
Mean interval eurve of the length of the leaves of 11 fertile stems of 
Mnium punctatum. 
Intervals: 1. 2. 3. 4. 5. 6. T. 8. 9. 10. 
Meanlength,in mm... 2:10 256 353 363 459 498 573 017 699 7:19 
Id. reduced in °/, 
of the length in | 29 36 49 50 64 69 80 86 97 100 
interval 10 ...... 
$8. PERCENTAGE OURVES.— Comparison between an interval curve of one 
specimen and a mean interval curve. A comparison between specimen curves 
of two or several stems * is only suitable when the number of leaves is the 
same (see $6); as this number is very variable, equality rarely occurs. 
Therefore we must have recourse to interval curves ($$ 6 & 7). 
Interval curves (specimen as well as mean curves) in which absolute values 
are given are all comparable with each other, because they have all the 
same number of ordinates. In such curves, however, the absolute values of 
the characters under consideration are of secondary importance ; the relative 
values of the successive ordinates, indicating the variation along the acis, are 
much more important (see also $ 14). Therefore we are allowed to bring the 
* Belonging to the same species or to difterent species. 
