It is evident, therefore, that bacterial growth in substances used as food is not necessarily injurious and may in some cases increase the palatability of food without destroying its wholesomeness. Little or nothing is known about the correlation of visible signs of decomposition with the presence of poisonous products, and it is at present impossible to say at what point in the process of decomposition a food becomes unfit to use owing to the accumulation of poisonous substances within it. There seems to be no connection between the natural repugnance to the use of a food and its unwholesomeness. Under ordinary conditions the nauseous character of very stale eggs is proverbial, and yet few nitrogenous foods have so clear a health record as eggs or have been so infrequently connected with food poisoning outbreaks. It might seem tempting to conclude on the basis of the available evidence that spoiled or decomposed foods possess poisonous qualities only when certain specific bacteria, like the _B. botulinus_ already discussed, have accidentally invaded them and formed definite and specific poisons. But we have no right to assume that the everyday decomposition products of the banal bacteria are in all cases without injurious effects. Even though no sharply defined acute form of poisoning may be laid at their door, it does not follow that an irritating or perhaps slightly toxic action of the ordinary decomposition products is altogether absent. Our present knowledge of the nature and degree of danger to be apprehended from the use of spoiled food is imperfect and unsatisfactory. That fact, however, does not release us from the obligation to continue measures of protection based even to a limited extent on experience. FOOTNOTES: [95] Another species of _Claviceps_ (_C. paspali_) which attacks the seeds of a wild grass is believed to be responsible for certain outbreaks of poisoning among cattle and horses (_Science_, XLIII [1916], 894). [96] Barger (_Jour. Chem. Soc._, XCV [1909], 1123) has shown that parahydroxyphenylethylamine is present in ergot and is in some degree responsible for the physiological action of the drug. [97] Although some of the early outbreaks were traced to the use of sausage, particularly in Württemberg, the proportion of recent botulism poisoning attributed to this food is no greater than of sausage-conveyed infections with the paratyphoid bacillus (chap. vi), and a number of the most completely studied outbreaks of botulism have been traced to ham, beans, and other foods. [98] _Deutsche Viertelj. f. öffentl. Ges._, XLV (1913), 8. [99] E. Sacquépée, _Progrès méd._, XXVI (1910), 583. [100] _Report to Local Govt. Board on Bacterial Food Poisoning and Food Inspection_, N.S. No. 77, 1913, p. 27. [101] _Southern Cal. Pract._, XXII (1907), 370. [102] _Ibid._, XXV (1910), 121. [103] _Arch. of Int. Med._, XIV (1914), 589. [104] _Amer. Med._, X (1915), 85. [105] _Jour. Amer. Med. Assoc._, LXI (1913), 2301.