140 Griffiths : Some Northwestern Ervsiphaceae 







the snow line^ which here is located at about 10500 ft, SpJiaero^ 

 thcca CastagJiei was found at tlie highest level on Amiica cordifoliai 

 This host IS very abundant here at lower altitudes, but this was 

 the only locality in ^\hich it was found affected with this fungus. 



Some attention has been paid to the distribution of the fungi on 

 closely related hosts of the region, A study of the list will furnish 

 the best general idea on this point. E. ciclioraccannn lias been 

 found, as usual, on a great number of hosts and in several instances 



ft 



widely separated ones. Wliilc its favorite habitat appears to be on 

 the composites it is by no means confined to this group. When 

 these facts are taken into consideration it is rather astonishing to 

 find an entire absence of the fungus on liosts wdiich are generically 

 related. A striking illustration came under my observation at 

 Sheridan, Wyo. In the corner of a garden tlircc species of the 

 genus Artemisia — A, Lndoviciana^ A. tridoitata and A. loigifolia 

 were growing in profusion. They were in such close proximity that 



I 



their branches were actually intertwined. Being in partially cul- 

 tiv^ated ground the growth of each was much more luxuriant than 

 usual. A. LiidoviciiDia was loaded with E, cichoraccaniDi, but care- 

 ful search failed to reveal any on the other two species. I am not 

 aware that the fungus has ever been recorded on either of these 



■ 



two species. This is all the more astonishing when we consider 

 the frequency with which A, Ludoviciana is affected. In the ab- 

 sence of direct experimental evidence no positive reason can be 

 given for the absence of this fungus on the two species in question 

 under such apparently favorable conditions. Although closely 

 related there is, however, a great difference in the aromatic prin- 

 ciple and the development of trichomes in the two species, which 

 may account for the phenomenon at least in part. A parallel case 

 was observed in two other species of tlie same genus at Buffalo, 

 Wyo. A. dracunculoidcs and A. Canadensis were growing together. 

 The former had an abundance of E, cichoraccanini upon it while 

 the latter was entirely free. A directly opposite effect even in 

 widely separated hosts w^as observed at Missoula, Mont. Here 

 Crataegus and Abuts were growing so that the branches over- 

 lapped. The former was badly affected with PJiyllaetinia suffnlta 

 and the latter with Microsphaera alni. A small quantity of P. 



m 



suffidta was found on the Abuts also. 



