850 



SCIENCE 



[N. S. Vol. XXXVIII. No. 



apply to the General Court of New HampsMre for 

 $1000 (tracing 14). Thereafter until the end of 

 the book the money is all in English pounds. 



We see in the above the gradual substitution of 

 the conventional $ sign for the spelled word. The 

 spellin? out of the word becomes less and less fre- 

 quent as the record proceeds. If we examine the 

 tracings of the signs, we find that the first eleven 

 have the S crossed by only one line. The last three 

 have the double line as it is used at the present 

 day. 



Florian Cajori 



Colorado College, 

 CoLOEADO Springs, Colo. 



A NON-CHROMATIC REGION IN THE SPECTRUM 

 FOR BEES 



To THE Editor of Science: The brilliant 

 work of Professor K. v. Frisch, of Munich, on 

 the color sense of bees (which follows upon his 

 very ingenious investigation of the color sense 

 of fishes and of crabs) seems to have been 

 strangely overlooked in this country, where 

 more confidence is placed in the very insuiE- 

 cient (from the point of view of logic) con- 

 clusions of Hess than they deserve, v. Frisch 

 carried on his experiments on bees in the ojien 

 air, in the close vicinity of an aviary; he 

 found that a single day's training was suffi- 

 cient to enable many hundreds of bees to form 

 the association : Whatever is blue is sweet, 

 whatever is gray (of any one of thirty-two 

 different shades) is not sweet. In the same 

 way they were able to learn, later, that yellow 

 indicates sweetness; no amount of training, 

 however (they were tried steadily for ten suc- 

 cessive days), could teach them to distinguish 

 between red and black. Training for green 

 had to be postponed for another year, on ac- 

 count of the oncoming of the cold and rainy 

 weather of autumn, which rendered the bees 

 too sluggish to carry on the work. 



Professor v. Frisch's results are so striking, 

 especially the proof of the total blindness to 

 red of his bees (shown already by Washburn 

 and by Watson in the case of higher animals) , 

 and his method (which I do not give here) 

 was so good — so convincing and so little con- 

 sumptive of time — that I was anxious to have 

 him, when the weather permitted, put to the 



test a question which had been in my mind for- 

 some time, namely, whether, when animals are 

 insensitive to red, there can not be found a 

 certain blue-green (its complementary color) 

 to which they are also insensitive — ^whether- 

 they have not, in other words, a dichromatic 

 (yellow and blue) color system only. I there- 

 fore wrote to Professor v. Frisch some three 

 weeks ago on this point, and I have now re- 

 ceived a reply from him. He writes me that 

 he has already tried the experiment, and that 

 my Vermuthung is justified. There is a com- 

 pletely non-chromatic region for the bee in 

 that part of the color-spectrum which corre- 

 sponds to blue-green for the normal eye: no 

 amount of training enabled the bees to pick 

 out this color from the series of grays, al- 

 though, as I have said, a single day sufficed to- 

 train them to alight, in hundreds, on yellow, or 

 on blue, and to leave the grays entirely un- 

 visited. This, combined with the fact that the 

 point of maximum brightness for bees is- 

 shifted well towards the green (the circum- 

 stance which led Hess to the erroneous con- 

 elusion that bees, as well as all other inverte- 

 brates together with fishes, are insensitive to- 

 chroma — that they have achromatic vision 

 only) shows in fact that their vision is 

 dichromatic instead of tetrachromatic, that 

 their colors are yellow and blue, and that 

 their vision resembles in type the protanopic 

 form of red-green blindness. 



That this quite extraordinary fact — ^the 

 non-specific quality to bees (as well as to 

 fishes) of the blue-greens — has not hitherto 

 been discovered by the investigators of the 

 color sense of animals is easy to understand, 

 for, since one can not readily try all the colors 

 of the rainbow, one naturally tries first the 

 " unitary " colors, red, green, yellow and blue, 

 instead of the " color-blends," blue-green, 

 yellow-green, red-yellow and red-blue (the two 

 last are popularly but most unscientifically 

 called orange and purple, respectively). One 

 forgets, what ought to be a perfectly familiar 

 fact, and would be were it not for the innu- 

 merable color-illusions which the Hering 

 color-theory forces upon its adherents, that 

 though the red-green blind individual never 



