792 



SCIENCE 



[N. S. Vol. XXXVII. No. 960 



parison came to the conclusion that it was a 

 species of Merychippus. A more thorough 

 restudy of the Miocene horses last summer 

 brought me to the conclusion that this tooth, 

 ■while certainly distinct from Hipparion, lies 

 somewhere near the border line between 

 Merychippus and Protohippus, but on which 

 side of the line I can not determine except 

 arbitrarily. The species is, therefore,- in fact 

 indeterminate generically, and a valid genus 

 can not be based upon it. Eippodon would, 

 however, stand as the type of a group includ- 

 ing Merychippus, Protohippus and Pliohippus 

 as contrasted with Hipparion and Neohip- 

 parion. In stratigraphic correlation of the 

 beds at Bijou Hill, where it was found, it 

 would be listed under the Protohippinae as 

 Hippodon speciosus gen. et sp. indet. 



3. Deinodon Leidy is determinable as to 

 family, but is not determinable generically, as 

 the genera of carnivorous dinosaurs are now 

 distinguished. The same is true of a whole 

 series of genera and species described by Leidy 

 and Cope from the Judith Eiver. The treat- 

 ment of types and referred specimens of these 

 genera by paleontologists as specifically dis- 

 tinguishable or identical has sadly misled Dr. 

 Peale in his recent discussion of the verte- 

 brate evidence as to the age of the Judith 

 Eiver beds, leading him to present as conclu- 

 sive evidence of identity in age a correspond- 

 ence in fauna which to those who know the 

 nature of the specimens on which the lists are 

 based is no evidence at all. 



In brief the plea is for the full recognition 

 of nomenclature laws, but for the avoidance of 

 arbitrary or unprovable identifications in the 

 future, and the recognition of the actual facts 

 as to the extent to which described genera and 

 species are truly determinable. The allowed 

 exception in the case of topotypes is based 

 upon an inference of identity which it would 

 seem impossible ever to prove incorrect. In 

 all other cases the chances that future discov- 

 ery may upset an arbitrary identification 

 should prevent its being used as a basis for 

 changes in nomenclature. 



The source of the present lamentable situa- 

 tion in nomenclature is that an excellent sys- 



tem of procedure, designed to settle unsettled 

 questions, has been wrenched from its intent 

 and used to unsettle settled questions. The 

 present writer, having studied with more or 

 less care the majority of the type specimens 

 of American fossil mammals and reptiles, has 

 abundant evidence at his command to upset 

 by a strict application of the accepted laws 

 and procedures, much of the present nomen- 

 clature, including many of the alterations 

 proposed in recent years upon grounds of 

 priority. But he has no intention of so mis- 

 using his opportunities, or of being respon- 

 sible for such changes until convinced that 

 they will really result in greater stability. 

 W. D. Matthew 



HOW IS THE WORD FOOD TO BE DEFINED? 



The query expressed in the title " How is 

 the word 'food' to be defined?" is suggested 

 by a restrictive usage of this word which is 

 rather prevalent in American text-books of 

 elementary botany, and which seems to have 

 originated among American plant physiolo- 

 gists. Presumably it had its birth in univer- 

 sity courses in botany where the arguments 

 for its use were given and understood, but 

 as it appears in the elementary texts, it in- 

 volves a marked inconsistency of thought and 

 expression for which no provision is made. 

 Since it represents a striking divergence from 

 the ordinary meaning of the term " food," it 

 deserves wider consideration, looking either 

 toward its general adoption, if desirable, or else 

 toward its discontinuance. 



The word food, according to its ordinary 

 connotation, is applied to any substance which, 

 when taken into the body of an organism, can 

 be used by that organism in the construction 

 of new tissue. Definitions of essentially this 

 content are to be found in the Century, Stand- 

 ard and Webster dictionaries. Using this 

 definition as a basis, we should consider as 

 food for green plants the water, carbon dioxide 

 and mineral salts absorbed from the surround- 

 ings. According, however, to the restricted 

 usage, these are not considered as " foods," 

 but are referred to as " raw ma'iorials," " nu- 

 trients," " food materials," or some other cir- 



