HAY FEVER 203 



ficity depends upon the chemical composition of the antigenic protein, 

 rather than its biological origin, for I have found it possible to dis- 

 tinguish in the hen's egg five distinctly different antigens, and these 

 correspond to five proteins which have been distinguished by chemical 

 measures."" Together with Dr. T. B. Osborne, working with purified 

 vegetable proteins, I have found evidence that a single isolated pro- 

 tein (hordein or gliadin) may contain more than one antigenic radical."" 

 As Osborne^^ has said, "chemically identical proteins apparently do 

 not occur in animals and plants of different species, unless they are 

 biologically very closely related." Whether the chemical differences 

 that determine specificity are of quantitative nature, which can be 

 disclosed by analytic means, or whether they are sometimes dependent 

 upon spatial relationships of the amino-acid radicals, as Pick sug- 

 gests, remains to be determined. My own experience indicates that 

 usually, at least, proteins distinguishable by anaphylactic reactions 

 also show readily distinguishable chemical differences. 



HAY-FEVER 



In 1902 Dunbar"- demonstrated conclusively that typical hay-fever, in its 

 several forms, is due to pollen of various sources; in all, twenty-five varieties of 

 grass and seven varieties of plants of other sorts being found whose pollen, when 

 placed upon the nasal or conjunctival mucous membranes of haj'-fever patients, 

 causes a typical attack of the disease. In Germany the disease seems to come 

 chiefly from pollen of the grasses and grains (rye pollen being most active), whereas 

 in America the most important pollen seems to come from members of the Ambrosia 

 (ragweed)"' and Granimaceae (grasses). Dunbar also found that the toxic con- 

 stituent could be dissolved from the pollen in salt solution, and seemed to be a 

 protein. 



The protein constituents of the pollen of rye ha\e been studied further by 

 Kammann,^^ who found three proteins, one of which, an albumin, was found to 

 contain all the active matter. This constitutes about 5.5 per cent, of the entire 

 weight of the pollen, is weakened but little by heating to S0°, and is not destroyed 

 by boiling; it is but partly destroyed by pepsin and trypsin, and resists acids but 

 not alkalies. Analysis of pollen from Ambrosia by J. H. Koessler"^ showed that 

 most of the nitrogen present was protein nitrogen, with 14.72 per cent, arginine, 

 14.05 per cent, histidinc and 3.18 per cent, l.ysine. Heyl'^ obtained from Ambrosia 

 pollen a mixture of albumose, peptone, albumin and glutelin, the latter being most 

 abundant. A solution containing O.OOS mg. of pollen protein, which amount is 

 contained in two or three pollen grains, produf^es a reaction in susceptible indi- 

 \aduals, but large amounts have no effect on normal persons. 



Dunbar manufactured an "antitoxic" serum by immunizing horses against 

 the pollen, believing that he was dealing with a toxin, but its efficacy is more than 



^' The chief proteins obtained by fractionating serum give cross reactions with 

 each other, but strongest \vith the homologous protein (Kato, Mitt. med. Fak. 

 Univ. Tokio. 1917 (IS), 195. 



"0 Jour. Infec. Dis., 1913 (12), 341. 



"' Harvey Lectures, 1910-11. 



'- Full re^^ew of subject and literature given by Prausnitz, KoUe and Wasser- 

 mann's Handbuch, 1913 (2), 1469; Koessler, Forchheimer's Therapeutics, 1914 

 (5). 671. 



"' See Cooke andVan der Veer, Jour. Immunol., 1916 (1), 201; Goodale, Boston 

 Med. Surg. Jour., 1914 (171), 695. 



"^ Hofmeister's Beitr., 1904 (5), 346; Biochem. Zeit., 1912 (46), 151. 



'5 Jour. Biol. Chem., 1918 (35), 415. 



"« Jour. Amer. Chem. Soc, 1917 (39), 1470; 1919 (41), 670. 



