398 Journal of Agricultural Research voi. ix t No.« 
opinion, from experiments in which E. ageratoides was fed to various 
animals, that this weed was the cause of the disorder. His results were 
not entirely convincing, however, and in criticism of them Crawford (i, 
p. 15) says: 
It can not be said that Moseley has even proved Eupatorium ageratoides to be a 
poisonous plant, much less the cause of "trembles/* 
When, in the summer of 1906, about 50 head of cattle died of trembles 
near Minooka, Ill., an investigation of E. ageratoides was undertaken by 
the Office of Poisonous Plant Investigations of the United States Depart¬ 
ment of Agriculture, since it was the popular belief that this weed was 
the cause of the trouble. Aqueous extracts were prepared from dried 
plants and plants preserved in water to which a small amount of chloro¬ 
form had been added. These extracts were fed or injected subcutane¬ 
ously into rabbits, cats, and dogs. An aqueous extract from the ash of 
dried plants was also administered to rabbits. Fifty-eight gm. of fresh 
plants were, furthermore, fed to a lamb weighing 25 kgm. without the 
production of symptoms of trembles. In summarizing the results of 
these experiments, Crawford says (1, p. 19-20): 
It certainly can not be said that it has been proved that milksickness is due to any 
constituent of Eupatorium ageratoides. * * *. Again severe epidemics have 
occurred in winter when the foliage has disappeared, which would tend to exclude the 
higher, nonevergreen plants as the cause of this disorder. In fact, all the evidence in 
hand is against the causation of this disease by such plants. 
Subsequent publications by Moseley (5, 6) advance the claim that 
aluminium phosphate found to be present in leaves and stems of E. 
ageratoides is the active toxic principle. Animals fed on this weed or on 
food in which aluminium phosphate was mixed were found to void 
aluminium phosphate in the milk and urine. The blood of these animals 
and certain organs were, moreover, found to contain aluminium phosphate. 
The stems of rayless goldenrod {Isocoma heterophylla) were also found to 
contain aluminium phosphate; and when this weed, too, was fed to rabbits 
it produced symptoms similar to those following the feeding of E. 
ageratoides or aluminium phosphate. Since aluminium phosphate is 
insoluble in water, this accounts, as explained by Moseley, for Crawford’s 
failure to produce poisoning in the experiments in which aqueous extracts 
of E. ageratoides were used. His criticism of the experiment in which 
58 gm. of fresh weed were fed to a lamb weighing 25 kgm. is that this 
quantity would probably not be fatal to a full-grown rabbit. 
Experiments conducted by Jordan and Harris (2, 3) in New Mexico 
and Texas, where this disease is present, but where E. ageratoides does not 
grow, indicate that the disease is of bacterial origin. Trembles developed 
in rabbits, guinea pigs, dogs, and calves by inoculation with a spore¬ 
forming organism which the authors described as Bacillus lactimorbi. 
This organism was present in the milk and butter of cows affected with 
trembles, in the feces of nonfatal cases in man, in certain parts of the 
