9 s 
W. B. Grove. 
approach to the ideal classification, but it is subject to the same 
fatal objection as Schroter’s. Arthur, for instance, separates P.fusca 
from P. Pruni-spinosce , placing it in his genus Polythelis, though he 
brings P. cohcesa and P. Pruni-spinosoe together. If a method of 
classification is only correct in so far as it coincides with the 
groupings that have originated in a line of descent, then no such 
method can be satisfactory that does not place these three species 
in a common genus. 
Dietel’s Rule. 
To illustrate Dietel’s “ Rule,” we may consider the numerical 
relations of the species of Puccinia and Uromyces (the two highest 
evolved genera of the Uredineae) to their hosts. The number of 
Dicotyledons in the world is to the number of Monocotyledons 
about as 5 : 1 ; the number of Pucciniae parasitic on Dicotyledons 
is to the number on Monocotyledons about as 2 : 1 ; in the genus 
Uromyces the proportion is about the same (23). In other words, 
the distribution of the parasites has not yet attained so great an 
extension on the higher class as on the lower. The evolution of 
the parasites naturally lags behind that of the hosts. 
The species of Puccinis on the Composite number over 300, 
i.e. nearly one-quarter of the known total. In Britain the number 
on Compositae is also about one-quarter of the British Puccinis. 
The species of Uromyces on Leguminoss number about 120 
out of over 500 in the world; i.e., again about one-quarter: the 
proportion in Britain is much the same. If the species of Uromy- 
cladium, all on Acacia , are included (and this genus differs from 
Uromyces mainly in bearing its teleutospores not singly, but in 
clusters upon a common stalk) then the world-proportion would be 
over one-quarter. The striking point is that the greatest number of 
each genus lives on orders belonging to the largest and most highly 
evolved sections of the gamopetalous and polypetalous groups 
respectively. Puccinia seems to have specialised upon the Com- 
positae in colder lands, and Uromyces upon the Leguminosae in 
warmer lands, for the species of Puccinia on the latter are hardly 
more than 1%, and of Uromyces on the former about 6%, of the 
totals. 
The number of Puccinice on the Grasses is about 150, and of 
Uromyces about 50; i.e., in each case in the neighbourhood of 10%. 
The relations of the numbers of these genera to other families are 
equally interesting, but would take too long to unravel here. It is 
worthy of mention, however, that there are few on the Rosaceae, 
