E. M. East and R. W. Glaser.
Observations on the Relation between Flower Color and Insects.
Psyche 21(1):27-30, 1914.

This article at Hindawi Publishing: https://doi.org/10.1155/1914/65740
CEC's scan of this article: http://psyche.entclub.org/pdf/21/21-027.pdf, 244K
This landing page: http://psyche.entclub.org/21/21-027.html

The following unprocessed text is extracted automatically from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

19141~ East and, Glaser-Relation Between Flower Color and Insects 27 segment, and the posterodorsal portion of the head is much less deeply and exten- sively infuscated.
Described from four specimens taken by Mr. T. Fukai in the Saitama division of Japan.
OBSERVATIONS ON THE RELATION BETWEEN FLOWER COLOR AND INSECTS.
BY E. M. EAST and R. W. GLASER.
Bussey Institution, Harvard University.
In 1909 a cross was made between the small red flowered Nico- tiana forgetiana Hort (Sand) and hTicoiiana alata Lk. and Otto var. grandiflora Comes, the large white N. affinis of horticulture, for the purpose of studying certain problems of heredity. About fourteen thousand plants of the second, third and fourth hybrid gen- eration have been grown, and it has been established beyond a rea- sonable doubt that each plant is completely self-sterile though it crosses easily with any of its neighbors. Several hundred carefully controlled self-pollinations have not yielded a single seed, while histological studies have shown self-fertilization to be practically impossible. On the other hand, hundreds of artificial cross pollina- tions have yielded capsules full of seed in almost every instance, showing with what ease cross-fertilization takes place, for artificial pollination is usually not as successful as natural pollination. The fact that every capsule formed naturally on these plants must have resulted from a cross-pollination produced by an insect, serves to excuse our adding to the already huge literature on the relations between insects and plants. The sixteen different color forms that have segregated from the original cross permit observa- tions on the percentage of flowers cross-fertilized and the selective value, if any, of distinct color varieties. Our knowledge of the behavior of insects relative to flowers has been greatly extended during the past few years by the work of Plateau, Forel, Lovell, Grsenicher and others, but it has resulted in that obscurity which precedes aggregation and precipitation by disclosing the marvelous complexity of the relation. The adjust- ment between certain insect forms and certain types of flowers is



================================================================================

28 Psyche [February
just as obvious now as when pointed out by Sprengel, but few entomologists or botanists will admit its adequate interpretation by the simple natural selection idea as believed by Hermann Muller and his followers who did not see the obstacles to this view as plainly as did Darwin.
The attitude of botanists has been affected chiefly by genetic investigation. Mendelian research and hypotheses regarding mutational evolution have at least gained a serious reconsideration of the origin, inheritance, and cause of survival of flower forms. Investigations on cross- and self-fertilization, by giving a clear and reasonable interpretation of the vigor of first generation hybrids and the converse-the apparent deterioration through inbreeding hybrids-have caused us to view mechanisms for cross- pollination at a new angle. Self-pollination gives inherently stranger races (vigor not masked by heterozygosis) and insures reproduction, but practically precludes the trial of variations not of decisive value or of various recombinations of new variations with old characters. On the other hand, cross pollination, while permitting the survival of weak types through the vigor of hetero- zygosis, and while rendering reproduction more dubious, does assure a trial of all new variations in all the combinations possible in a mendelian sense.
The appreciation of the intricacy of the behavior of insects to- ward flowers is due primarily to the knowledge of insect sense or- gans, to the ingenuity of the experiments of animal psychologists, and to the passing of the tendency to interpret all the actions of the lower animals as tropisms.
For these reasons the question as to whether particular flower colors have a survival value due to the preference of certain insects for them, upon which we have gathered a few data, would probably be answered somewhat as follows by the majority of biologists. Excluding any question of olfactory sense, it may be assumed that insects perceive color differences from short distances but seldom if ever exercise a choice. Night flyers, of course, perceive white much more easily than colors. These conclusions are supported by the data in the following table:




================================================================================

19141 East and Glaser-Relation Between Flower Color and Insects 29' Flower color.
White. .........
Yellow. .........
Red ............
Purple ..........
Total number
of flowers on
10 average plants.
18,035
26,686
14,165
9,721
Total number
of flowers
fertilized.
Per cent. of
flowers
fertilized.
Ten average plants of each of the four colors-white, yellow, red and purple-were selected at random.
The total number of
flowers produced on each color type during the flowering season (July 15 to October 15) was determined by counting the places on the racemes where flowers had been. The number of capsules present was assumed to be the number of flowers fertilized, al- though this count is not as accurate as the first by reason of the accidental loss of capsules. Long experience with Nicotianas, however, leads us to believe that this error is small. The first point to be noted is the comparatively small percentage of cross-pollination by insects. Numerous experiments on artificial cross-pollination have shown that a very small amount of pollen causes normal development of the capsules, yet the yellow, red and purple types had only about 17 per cent. of their blossoms crossed. According to the table, the percentage of white flowers fertilized was more than twice as high as any of the colored types. The reason for this is obvious. From the beginning of the flower- ing period, about July 15, to the end period of summer heat, about September 15, the flowers opened at about 4 p. m. and remained open until about 7.30 a.m. During the last month of flowering, the weather was so cool that the flowers also were open throughout the day. Nearly two thirds of the fertilizations occurred during the last month as could be determined by the positions of the flowers on the racemes. Furthermore the percentage of fertiliza- tions on the white type during the last month was about the same as on the colored types. Roughly, one might say then that about 6per cent. of the pollinations of the colored types were made by night- flyers (Sphingidse, etc.), while during the same period these insects pollinated from 20 per cent. to 25 per cent. of the white type. In



================================================================================