Igto] CURRENT LITERATURE 235 
has been brought out by FiscHER" in the case of the gymnosporangia inhabiting 
Juniperus. A casual attempt to infect Sorbus aria, S. aucuparia, and some other 
species, with teleutospores obtained from Juniperus communis and ee “ 
be those of Gymnosporangium tremellotdes, resulted in no infections. This e 
ment showed that the teleutospores were neither those of G. tremelloides nor a 
of G. juniperinum, whose aecidial host is Sorbus aucuparia. These results led to 
a series of cross infections, in which the teleutospores of the Gymnosporangium 
were used to infect a number of pomaceous plants, and aecidial spores from these 
plants were sown on Juniperus. The work showed that two forms of Gymno- 
sporangium occur on Juniperus communis and J. nana. The first form occurs. 
chiefly on the stems and branches, rarely on the leaves, and has its aecidial form 
on Amelanchier ovalis. The first aecidia appear on A, ovalis 55 to 60 days after 
the sowing of teleutospores, and the teleutospores appear on Juniperus about 
20 months after the sowing of aecidiospores. For this form the specific name 
Amelanchieris, originally applied to the Roestelia on Amelanchier, is reserve 
The second form occurs chiefly on the leaves of Juniperus, and has its social 
form on Sorbus aucuparia, but does not infect A. ovalis. The aecidia appear 
45 to 50 days after the sowing of teleutospores, and teleutospores are produced 
in the spring following the sowing of the aecidiospores in late summer. This is 
the form known as G. juni perinum. 
For a discussion of the history, nomenclature, and geographical distribution 
of the two forms, as well as of their respective hosts, the reader is referred to the 
original paper.—H. HassErprinc. 
Fermentation of tobacco.—Theories relating to the fermentation of tobacco 
have been based either on the view that the process is due to bacterial action, or 
to the action of enzymes without the intervention of bacteria. One investigator 
alone, SCHLOESING, seems to have believed that the later part of the fermentation 
is due to purely chemical oxidations. In view of these conflicting theories describ- 
ing the process, the experiments of BorkHour and DEVrigs are especially inter- 
€sting.‘7 These authors attacked the problem by methods similar to those used 
in their recent work on the spontaneous heating of hay. Dry samples of cured 
tobacco were sealed up in tubes with oxygen or air, and in some cases water was 
added to the samples. The tubes were then heated to 100° or to 33°, according 
to the purpose of the experiment, and after a’definite period of heating the gas in 
the tubes was analyzed. It was found that oxidation took place at both tempera. 
tures, but more rapidly at the higher temperature. The presence of water 
increased the rate of oxidation. By the process of oxidation carbon dioxi 
liberated and oxygen was fixed. It was found that starch and pentosans were 
cea 
*° FiscHer, Ep., Studien zur Biologie von later ieee juniperinum. Zeit. 
Bot. 1: NS hes 2 + ko 
? BoeKnour, F. W., anp <table O. J. J., Ueber Tabaksfermentation. Cen- 
tal Bake M : 24: 496-511. 1909. 
