INTRODUCTION. 3 



the last research of Fayrer and Lander Brunton until the authors of this paper 

 resumed the work in 1882. One of them (Dr. Mitchell) hud long felt that it 

 would be well to revise the toxicology of our American serpents which he had 

 begun in 18o8, and as the later English observers had in some points differed from 



Fig. 1. 



him, to learn if they or he were correct, or whether the divergence as to results 

 was due to variations in the qualities of the venoms employed. Then too he had 

 become conscious of certain errors in his former researches, and wished to aid in 

 correcting them, and in rilling up some of the gaps left in this branch of toxi- 

 cology by himself and others. 



The authors started with a theory long held by Dr. Mitchell that snake venoms 

 are not simple in composition, but composed of two or more poisonous substances, 

 and that in the qualities and quantities of these agents would be found an expla- 

 nation of the differences between serpent venoms as to power to kill and mode of 

 causing death. 



How fertile has been the germinal idea of this research must be judged of by 

 this present essay; which will, we trust, by leading thought and experiment in 

 new directions hasten the day when we shall be able to treat with success the 

 wretched thousands who now perish annually by snake-bite in India and elsewhere. 



Some of our earlier results were so soon talked of and even noted in public 

 prints, that it seemed wise for this, and all other reasons, to state what we then 

 knew. This was done in a " Preliminary Report to the United States National 

 Academy of Sciences, in April, 1883." In this brief essay we announced our 

 proofs of the complex nature of snake poisons. The report was incomplete, and 

 in the light of our present more elaborate essay may be seen to contain several 

 erroneous statements. 



It is not in the nature of things, that a research along such varied lines as 

 our present volume follows, though extending over several years, should be per- 

 fect in detail, or complete for all genera of Thanatophidians. It is our earnest 



