94 
W. B. Grove. 
closely connected cases, e.g., P. Festucce Plowr., with its secidia on 
Lonicera, and P. Beckmannice, as well as species of Uromyces 
(coronatus, phyllodiorum, etc.)- Nevertheless, the series of forms 
given above on Rhamnus and grasses bears a close similarity to 
that on Anemone and Primus. If the same explanation is to apply, 
one must remember that the Graminese (and also the Cyperaceae) 
are not, as has sometimes been thought, primitive families but 
more recently evolved ones, being derived like the Orchidaceae 
from the Liliaceous type, though by different routes. 
We find much the same relations existing in other groups of 
species; not to labour this point, I will refer briefly to two alone. 
The well-known P. graminis has its aecidia and spermogones on 
Berberis, and its uredo- and teleutospores on grasses all over the 
world. P. Berberidis Mont., from Chile, has aecidiospores and 
teleutospores on Berberis; P. Meyeri-Alberti Magn. from Chile and 
Patagonia, also on Berberis, has probably teleutospores only, which 
differ from those of P. Berberidis merely in having a greater 
length. The aecidia of P. Berberidis are not unlike those of 
P. graminis, but in the former case teleutospores are produced 
round the aecidial cup, obviously from the same mycelium (6.) There 
is also P. Arrhenatheri Erikss. (from Europe and Western Asia) 
which has its spermogones and secidia on Berberis and its uredo- 
and teleutospores on Arrhenatherum elatius. The teleutospores of 
this species are similar to those of P. graminis, but on Berberis its 
perennial mycelium causes witches’-brooms, while the aecidium of 
P. graminis does not. Mention must also be made of AEcidium 
Maghellanicum and AB. Jakobsthalii-Henrici, both on Berberis and 
from the antarctic regions, of which AE. Maghellanicum causes no 
witches’-brooms, but the other does; their teleutospores are unknown. 
All these may be descendants of one ancestral autoecious species 
which grew on Berberis and of which the uredo- and teleutospore 
phases migrated to the Graminese, again without violating 
Dietel’s “ Rule.” 
The second instance is that ofP. Caricis, which has its aecidium 
on Urtica and its teleutospores on Carex. Here also we find a 
Micropuccinia on Urtica (P. Urticae Barcl., from Simla) which 
has teleutospores only, similar to those of P. Caricis but occurring 
in dense clusters on coloured spots like those made by the 
mycelium of the aecidium of the latter on Urtica, and therefore, in 
a manner quite unlike the teleuto-sori of P. Caricis. In this 
connection it may be as well to call attention to the fact that, in 
many cases, the form of the sori and the colouring of the spots on 
