l62 REPORTS ON INVESTIGATIONS AND PROJECTS. 



cies of corals will endure atmospheric exposure on a glass plate in the shade 

 for half an hour without apparent damage; nearly all will stand an hour 

 without harm, while some will stand 4 hours' exposure under the conditions 

 stated. Favia fragum, Pontes clavaria, and P. astreoides have the greatest 

 capacity for withstanding atmospheric exposure, while that of Mceandra 

 areolata and Siderastrea radians is almost as great. A number of species 

 withstood exposure on a glass plate in the sun for i]/ 2 hours, the specimens 

 being badly damaged, but not entirely killed. 



The studies of growth-rate were continued in 1912 as in 191 1. There are 

 under observation (1) colonies reared from planulae that settled in the 

 aquarium; (2) colonies from planulae that attached themselves to collectors; 

 (3) specimens cemented to tiles; (4) colonies growing naturally at one of 

 four stations. No detailed statement of these experiments and observations 

 has been made, except accounts of the rearing experiments, tables giving 

 the size of the yearling colonies, and three plates illustrating a few planted 

 corals, have been published in the Year Books of this Institution. As there 

 are many hundred growth records, it is not possible to present them here, 

 but they will be published in the next series of papers from the Tortugas 

 Laboratory, and a summary of the results to date will be given. The experi- 

 ments have been very successful and will furnish accurate information on 

 the growth-rate of nearly all the species of corals found in the Florida reef 

 region. 



Report on the Limits of the Spectrum for Birds, by John B. Watson, of 



Johns Hopkins University. 



While at Tortugas I began a study of the limits of the spectrum in day 

 birds. This work was begun primarily for the purpose of testing the theory 

 which has been advanced often to explain the homing sense. Certain French 

 investigators assume without experimental proof that homing birds use for 

 orientation rays in the spectrum which lie in the infra red. They assume 

 further that since these rays (heat rays) follow the curvature of the earth, 

 the birds without rising to extraordinary heights may directly perceive the 

 goal (cote, nesting locality, etc.). While this theory seems more or less 

 absurd upon its face, it deserves at least an experimental test. Entirely apart 

 from its bearing upon the homing sense, data upon the visible spectrum 

 of animals are much needed in comparative psychology and in sensory 

 physiology. 



A large spectrometer was constructed, which enabled the experimenter to 

 select any monochromatic bundle at will. The selected spectral band was 

 allowed to fall upon a plaster surface. 



The birds were tested in an experimental cage, as follows : 



A home box gave entrance to two compartments. These two compart- 

 ments were separated by a partition. Each compartment contained a plaster- 

 paris surface for the reception of the light. Only one compartment was 

 illuminated in any one test. The animal when released from the home box 

 might choose either the lighted compartment or the dark one. It was fed if 

 it chose the lighted compartment, but was not fed if it chose the dark one. 

 The apparatus was so arranged that the lighted compartment could be 

 offered either on the right or left side. It is clear that after an animal has 

 been trained in this way it will continue to respond as long as the plaster- 

 paris surface is illuminated by rays which affect the receptor. 



