August 19, 1904.] 



SCIENCE. 



245 



that reference should be made to the fact that 

 the first map of this character published in the 

 United States was constructed by Dr. Wm. 

 McMurtrie and published in a special report, 

 No. 28, of the U. S. Dept. of Agriculture in 

 1880. In Dr. McMurtrie's map the area of 

 ■ successful beet culture was indicated by a 

 region contiguous to a line showing a tem- 

 perature of 70° Fahrenheit during June, July 

 and August and rainfall above two inches 

 for the same months. 



The remarkable success which attended the 

 growth of sugar beets in the semi-arid regions 

 of California in the early '80's led me to dis- 

 card the factor of rainfall and to construct a 

 map based upon temperature conditions alone, 

 during the growing season. Some of the very 

 best areas in the United States, suitable for 

 beet culture are in arid and semi-arid regions. 

 The idea, however, of mapping the beet areas 

 in this country should be credited to Dr. 

 McMurtrie. H. W. Wiley. 



JORDAN ON FOSSIL LABROID AND CH^TODONT 

 FISHES. 



In a recent paper on Eocene fishes from 

 Monte Bolca, the present writer followed Dr. 

 D. S. Jordan in his excellent ' Review of 

 Labroid Fishes ' (Bept. U. S. Fish Comm. for 

 1887 (1891), pp. 559-699) in uniting Cuvier's 

 genus Grenilabrus with the earlier described 

 Symphodus of Rafinesque. He also adopted 

 the views of Agassiz and most recent authors 

 in referring Pygceus to the Chaetodonts. Since 

 this paper was published,* the writer has re- 

 ceived from President Jordan some valuable 

 notes on the relations of the Eocene forms to 

 existing genera, and as these are of extreme 

 interest to students of both fossil and modem 

 faunae, advantage is here taken of his per- 

 mission to make them available for others. 



The following extracts may be considered 

 as the most recent and authoritative opinion 

 that has been pronounced upon the afiinities 

 of the forms in question. 

 Deab Dr. Eastman: 



THe following notes are at your disposal for any 

 use you may care to make of them. I should now 

 separate Grenilabrus C. & V. as a valid genus 



* Bull. Museum Comp. Zool., Vol. XLVI., No. 1, 

 June, 1904. 



from Symphodus Raf. {^Corious C. & V.), with 

 which I united it in 1891. Symphodus (scina) 

 has the general characters of Grenilabrus, the ser- 

 rated preopercle and other features, but it has the 

 snout strongly produced, giving a concave profile, 

 a matter probably worthy of generic distinction. 

 Grenilabrus like Symphodus has 13 to 15 dorsal 

 spines, and 31 to 33 vertebrae. I do not see how 

 C sizajnochw can be properly placed in it, as these 

 numbers are fairly constant within the same 

 genus. C. szajnochce should form the type of a 

 new genus. 



Pygceus appears to be a generalized type, with 

 traits ancestral to the Chjetodonts, the Acan- 

 thuridae, and to /St(70K.Ms (Amphacanthus) . It has 

 the long and strong dorsal spines and the numer- 

 ous anal spines of Siganus, as well as the few soft 

 rays, large vertebrae and the general form of the 

 body. The teeth are apparently much as in 

 Siganus, and not brush-like as in the Chsetodon- 

 tidae. The Siganidse are scaleless, but in Pygceus 

 the scales are small and prickle-like as in the 

 Acanthuridae and Zanelidae. 



The ventral fins in Pygceus seem normal (I., 5), 

 but in Siganus the inner ray is also a strong spine, 

 the formula being I., 4, I. In most Chtetodonts 

 there are but three anal spines, and no more in 

 any of the Acanthuridae. Megaprotodon, a genus 

 of Chaetodonts, has, however, four anal spines, and 

 its form suggests resemblance to Pygceus. 



Taking the somewhat minute differentiation of 

 families accepted amongst recent fishes, Pygceus 

 may be made the type of a family Pygaeidae, allied 

 to Siganus on the one hand, and to Megaprotodon 

 and Zanclus on the other; and, perhaps, related 

 to the common ancestor of all these. * * * 



After Gill I use the name ' Teuthis L.' for 

 Teuthis hepatus, for which Linnaeus devised the 

 name. This is equivalent to ' Acanthurus ' of 

 most authors. ForskSls's original Acanthurus 

 (=^Monoceros, ^^Haseus) .has two hooks on the 

 tail instead of a sharp spine, the ventrals being 

 I., 3, instead of I., 5. Acanthurus gaudryi will 

 probably turn out a new genus near Teuthis, but 

 not quite the same. In our day Teuthis has nine 

 dorsal spines and a movable caudal spine. A. 

 gaudryi is nearer the Pacific genus Zebrasoma, 

 which has three to five dorsal spines and the soft 

 rays high. * * * 



David S. Jordan. 



It should be noted that Agassiz suggested 

 the propriety of dividing Pygceus into two 

 groups, P. iolcanus being taken as type of the 

 one, and P. coleanus of the other. President 



