=Arsenic.=--Arsenic is probably the most popular of the rat poisons, owing to its cheapness, yet our experiments prove that, measured by the results obtained, arsenic is dearer than strychnin. Besides, arsenic is extremely variable in its effect upon rats, and if the animals survive a first dose it is very difficult to induce them to take another. Powdered white arsenic (arsenious acid) may be fed to rats in almost any of the baits mentioned under barium carbonate and strychnin. It has been used successfully when rubbed into fresh fish or spread on buttered toast. Another method is to mix twelve parts by weight of corn meal and one part of arsenic with whites of eggs into a stiff dough. An old formula for poisoning rats and mice with arsenic is the following, adapted from an English source: Take a pound of oatmeal, a pound of coarse brown sugar, and a spoonful of arsenic. Mix well together and put the composition into an earthen jar. Put a tablespoonful at a place in runs frequented by rats. =Phosphorus.=--For poisoning rats and mice, phosphorus is used almost as commonly as arsenic, and undoubtedly it is effective when given in an attractive bait. The phosphorus paste of the drug stores is usually dissolved yellow phosphorus, mixed with glucose or other substances. The proportion of phosphorus varies from one-fourth of 1 per cent to 4 per cent. The first amount is too small to be always effective and the last is dangerously inflammable. When homemade preparations of phosphorus are used there is much danger of burning the person or of setting fire to crops or buildings. In the Western States many fires have resulted from putting out homemade phosphorus poisons for ground squirrels, and entire fields of ripe grain have been destroyed in this way. Even with commercial pastes the action of sun and rain changes the phosphorus and leaches out the glucose until a highly inflammable residue is left. It is often claimed that phosphorus eaten by rats or mice dries up or mummifies the body so that no odor results. The statement has no foundation in fact. No known poison will prevent decomposition of the body of an animal that died from its effects. Equally misleading is the statement that rats poisoned with phosphorus do not die on the premises. Owing to its slower operation, no doubt a larger portion escape into the open before dying than when strychnin is used. The Biological Survey does not recommend the use of phosphorus as a poison for rodents.