344 



SCIENCE 



[Vol. LVI, No. 1447 



After seven days one plant of Brittle Wax 

 in the field cage reacted similariy to the 

 plants in the greenhouse. When the sec- 

 ond pair of leaves appeared, mosaic markings 

 were not apparent and it was thought that 

 weather conditions had checked the progress 

 of the disease in the plant. However, upon 

 examining the plants one week later, two plants 

 of the three upon which the insects had 'been 

 placed had developed typical mosaic symp- 

 toms. All of the plants in the check cages, as 

 well as in the twenty-foot row outside of the 

 cages, have remained healthy up to this time, 

 one month from the time of the appearance of 

 the first pair of leaves. 



The species of aphid used in these experi- 

 ments has been identified by Miss Eugenia 

 McDaniel of the entomology department as 

 Macrosiphum solanifolii. This species has been 

 collected on beans at other times, especially 

 early in the season. Phaseolus vulgaris is one 

 of the known hosts of this very polyphagous 

 species. 



SUMMARY 



The spread of bean mosaic was observed 

 during 1921 under conditions which strongly 

 suggested transfer by insects. The sudden ap- 

 pearance of the disease in the water cultures of 

 beans growing in the greenhouse, and infested 

 with aphids, indicated even more definitely the 

 mode of dissemination. Definite proof of the 

 transfer of the virus by Macrosiphum solani- 

 folii was obtained under controlled conditions, 

 both in the greenhouse and in the field. 



Eay Nelson 

 Department of Botany, 



Michigan Agricultural College 



THE EXTENSION OF THE X-RAY INTO THE 

 ULTRAVIOLET SPECTRUM 



It was found that when thermions liberated 

 from a tungsten filament were accelerated and 

 allowed to impinge on a metal grid maintained 

 at a variable positive potential, secondary 

 electrons were emitted from the grid. The 

 numlber of such secondary electrons emitted 

 were measured by means of a galvanometer 

 in series with the grid and a plate maintained 

 at a constant positive saturation potential. 



On plotting the secondary current as a func- 



tion of the accelerating voltage, acting on the 

 primary electrons, a sudden change in the 

 slope of the curve occurring at critical poten- 

 tials was interpreted in the usual way. The 

 energy-quantum relation V(volts) L (A) = 

 12320 was used to compute the equivalent wave- 

 lengths. 



The following table gives the preliminary 

 results thus far discovered. The quantities 

 bracketed are still doubtful. Those preceded 

 by an (o) are not found by the usual breaks in 

 the curves but are positions on the continuous 

 curves where the ratio of the number of sec- 

 ondary (s) electrons per primary (p) were 

 such as indicated in the s/p colnmn. At pres- 

 ent it appears as if the convergence wave- 

 length (s/p == 3) for tungsten ended at 91.2 A 

 and is followed by an absorption band extend- 

 ing probably down to 14 A. This is then fol- 

 lowed by the Ma line, here extrapolated as 

 7.04 A from the above measurements. 



TUNGSTEN 



Otto Stuhlman, Jr. 

 The University of North Carolina, 



