Januaet 28, 1916] 



SCIENCE 



135 



has often been difficult to iind the larvae in 

 healthy trees, although they were present in 

 greatly increased numbers in the cankers. 

 Thus this insect, which is undoubtedly of im- 

 portance as a carrier of spores to healthy trees, 

 would, as the infections grew old, become less 

 so owing to its increasing tendency to breed in 

 and frequent diseased trees to the exclusion 

 of healthy ones. Other insects which come in 

 this category are several species of moths of 

 the genus Sesia, although, in the case of these, 

 observations indicate that adaptation to a life 

 in cankerous tissue has not developed to so 

 great an extent. 



Of more importance in providing for the 

 spread of the chestnut bark disease are the 

 fresh wounds made by certain insects through 

 the outer bark of the tree to the cambium 

 whereby spores disseminated in various ways 

 ■can gain entrance. Of first rank among in- 

 sects which work in this way are Lepiura 

 nitensj which, as stated, is found in 90 to 95 

 per cent, of the trees over 10 inches in diam- 

 eter throughout the chestnut range, and the 

 •chestnut bast-miner Ecioedemia phleophaga 

 Busck, which is found abundantly in 95 per 

 cent, of the saplings and younger trees 

 "throughout the natiu'al range of the chestnut. 

 Less abundant, though also widely distributed, 

 are three species of Sesia — S. castanea Busck, 

 8. sciiula Harris, and 8. pictipes G. & R. All 

 these insects attack perfectly healthy trees and 

 make wounds at various situations over the 

 entire bark surface of trees from those of sap- 

 ling size to those which are matured. Further- 

 more, most of them are abundant and widely 

 distributed. These wounds are all holes made 

 by the larvae either for the extrusion of frass — 

 in which case they are present and used 

 throughout the entire larval life — or for exit 

 when the larvae are preparing to leave the 

 open to the cambium and surrounded by the 

 moist dead tissue necessary^ for the growth of 

 the spores. Thus practically all chestnut trees 

 in their natural range have numerous open 

 "wounds whereby wind-blown and rain-washed 

 spores can gain entrance. Young cankers 



sEankin, W. H., "Phytopathology," Vol. 4, 

 J). 242, 1914. 



have been found starting in wounds of both 

 types mentioned. 



Wound makers of another class are the 

 cicadas, tree crickets, tree-hoppers and aphides. 

 These puncture the bark in ovipositing or feed- 

 ing. In numerous eases the blight has been 

 found starting in cicada and tree-hopper 

 wounds. A possibility exists that these insects 

 both carry the spores and directly inoculate 

 the wound; but such a chance is slight from 

 the fact that insects of this kind normally fre- 

 quent healthy trees. 



In conclusion it may be said that in view of 

 the facts established, namely, that ascospores 

 are carried about by the wind in great num- 

 bers and that the pycnospores are washed down 

 the trunks by the rain, the role played by in- 

 sects in the transmission of this disease in 

 merely transporting the spores is insignificant. 

 On the other hand, owing to these same char- 

 acteristics of the disease, the part played by 

 insects in making wounds in the living cam- 

 biiun where such spores can gain entrance is 

 far more important, and such wounds have 

 been commonly found infected. Again, the 

 fact that certain insects frequent diseased 

 trees and eat the pustules, thereby preventing 

 the dissemination of the spores, can certainly 

 not be considered other than a benefit. 



F. C. Craighead 

 Bureau op Entomology, 

 U. S. Depaetment of Ageicultuee 



CANCER AND HEREDITY 



In final reply to Dr. Little, of Harvard, 

 there are three things to be said: 



My results in cancer transmission are these : 

 1. The establishment of strains of mice 

 which both in inbreeding and in hybridization 

 transmit spontaneous cancer through as many 

 generations as I please to carry it, and in a 

 percentage which can be predicted with a sur- 

 prising nicety. For example, certain strains 

 of mice have been producing a fairly steady 

 percentage of spontaneous cancer in this labo- 

 ratory for five years without a break. 



