The flicker response contour for the isopod Asellus is a simple probability integral ( F - log I ) over the whole determinable range ( F = 1 to 51). This contrasts with the "distorted" asymmetrical curves obtained with Apis, Anax , and other arthropods with large convex eyes. The explanation of the distortion as due to mechanical conditions affecting photoreception is therefore confirmed, as the structure of the Asellus eye does not make such a factor likely to be expected for this case. The Asellus curve agrees with the only other available complete and uncomplicated flicker response contour (from Pseudemys , turtle with rod-free retina), in showing the superiority of the probability integral formulation as compared with certain others which have been suggested. It is noted as a curious and probably important fact that the relative dispersion of the intensity thresholds (σ' log I ) for the elements implicated in determining the flicker contour appears to be identical in bee, dragon fly nymph, and isopod. Other relevant information derived from similar experiments with vertebrates shows that this quantity is specifically determined by the organization of the animal. The nature of the common feature of neural organization in three such diverse arthropods, as contrasted with the diversity seen within one class of vertebrates ( e.g ., teleosts), remains to be discovered.

1. At constant temperature, with a fixed proportion of light time in a flash cycle (namely, t L /t D = 1), the mean critical intensity for motor response to visual flicker by the turtle Pseudemys scripta follows a probability integral (log I ) as a function of flash frequency F . The fit is close and satisfactory; certain quite minor but consistent deviations are adequately explained by features of the experiments. 2. The variation (σ I ) of critical I is directly proportional to the mean critical intensity ( I m ), over the entire explorable range. 3. These facts are consistent with the fact that the retina of this turtle is devoid of rods. It contains only cones, histologically, which, with their central representations, provide a single population of sensory effects. The properties of this population are compared with those of homologous populations deduced from corresponding measurements with other forms (various fishes; amphibian; man) which exhibit two such groups of sensory effects associated with the possession of retinal rods and cones. 4. Certain other formulations which have previously been applied to homologous data obtained with other organisms do not properly describe the Pseudemys measurements. 5. The use of a probability integral to describe the data of response to visual flicker for the dissection of the compound curves provided by animals possessing both rods and cones, is accordingly Justified. 6. Persisting differences among individuals of Pseudemys as regards the values of the critical flash intensity under various conditions of experimentation are of the same order of magnitude as are the transitory differences found in lots of other kinds of animals. 7. Determinations of mean critical flash frequency ( F m ) at fixed levels of I lie slightly above determinations of I m at fixed values of I , as with other forms. The variation of critical flash frequency goes through a maximum as log I is increased; its height is lower than with certain other forms, in correlation with the low general slope of the F - log I curve (more properly, band). 8. These facts are consistent with the view that the dispersions of the individual critical intensities (and flash frequencies) are determined by organic variation rather than by "experimental error." 9. When the temperature is altered the F - log I m curve is shifted, with no change of F max. or of shape; the curve moves to lower intensities as the temperature is raised. 10. The reciprocal of the mean critical intensity, at fixed flash frequency, is a measure of excitability. With increase of temperature (12.5° to 36°) 1/ I m for given F follows the Arrhenius equation, exhibiting a "break" at 29.5° (µ = 26,700, 12.5° to 29.5°; 12,400, 29.5° to 36°). This is explained by the necessary theory that, the number of elements of sensory effect required for the index response at fixed F being constant, the ease of their excitation is governed by temperature through its control of the velocity of an interrelated system of catalyzed processes common to all of the sensory elements concerned.

Flicker response curves have been obtained at 21.5°C. for three genera of fresh water teleosts: Enneacanthus (sunfish), Xiphophorus (swordtail), Platypoecilius (Platy), by the determination of mean critical intensities for response at fixed flicker frequencies, and for a certain homogeneous group of backcross hybrids of swordtail x Platy (Black Helleri). The curves exhibit marked differences in form and proportions. The same type of analysis is applicable to each, however. A low intensity rod-governed section has added to it a more extensive cone portion. Each part is accurately described by the equation F = F max. /(1 + e - p log - p log I/I i ), where F = flicker frequency, I = associated mean critical intensity, and I i is the intensity at the inflection point of the sigmoid curve relating F to log I . There is no correlation between quantitative features of the rod and cone portions. Threshold intensities, p, I i , and F max. are separately and independently determined. The hybrid Black Helleri show quantitative agreement with the Xiphophorus parental stock in the values of p for rods and cones, and in the cone F max. ; the rod F max. is very similar to that for the Platy stock; the general level of effective intensities is rather like that of the Platy form. This provides, among other things, a new kind of support for the duplicity doctrine. Various races of Platypoecilius maculatus , and P. variatus , give closely agreeing values of I m at different flicker frequencies; and two species of sunfish also agree. The effect of cross-breeding is thus not a superficial thing. It indicates the possibility of further genetic investigation. The variability of the critical intensity for response to flicker follows the rules previously found to hold for other forms. The variation is the expression of a property of the tested organism. It is shown that, on the assumption of a frequency distribution of receptor element thresholds as a function of log I , with fluctuation in the excitabilities of the marginally excited elements, it is to be expected that the dispersion of critical flicker frequencies in repeated measurements will pass through a maximum as log I is increased, whereas the dispersion of critical intensities will be proportional to I m ; and that the proportionality factor in the case of different organisms bears no relation to the form or position of the respective curves relating mean critical intensity to flicker frequency. These deductions agree with the experimental findings.

In phototropic tests with young Limulus , the phototropic reactions to flickering fields were studied. If the two fields are equal in area and brightness but different in flicker frequency, the number of animals going to the two fields is proportional to their flicker frequencies. Equal stimulating effects of two fields differing in flicker frequency are obtained by reduction of the area of the faster flickering field. The areas for equal effect must be inversely proportional to their flicker frequencies. It seems that equal effects are dependent upon equality of the number of active excitation elements per unit of time.

Bees were conditioned to collect food on natural and artificial flower beds, parts of which could be set into rotation or side to side movement. Through the relative motion of the flowers the number of alternating stimuli upon the bee's eye is increased. Due to the fact that bees show a strong reaction to intermittent optical stimulation, the proportion of bees settling on the moving section of the flower bed is increased. It seems probable therefore that the visual reaction of bees to flowers in nature is largely due to the flicker effect produced through the motion of the bees relative to the flowers.

The curve of mean critical illumination ( I m ) for response to flicker as a function of flicker frequency ( F ) for the larvae of the dragonfly Anax junius is progressively shifted toward higher intensities the lower the temperature. The maximum flicker frequency (one half the cycle time of light and no light) and the maximum intensity with which it is associated are very little if at all affected by change of temperature. These facts are in agreement with the requirements of the conception that recognition of critical illumination for reaction to flicker involves and depends upon a kind of intensity discrimination, namely between the effects of flashes and the after effects of these flashes during the intervals of no light. The shift of the F-I m curve with change of temperature is quite inconsistent with the stationary state conception of the determination of the shape of the curve. The dispersion ( P.E. I I 1 ) of the measurements of I 1 is directly proportional to I m , but the factor of proportionality is less at high and at low temperature than at an intermediate temperature; the scatter of the values of P.E. I I 1 is also a function of the temperature. These facts can also be shown to be concordant with the intensity discrimination basis for marginal recognition of flicker.

The curve connecting mean critical illumination ( I m ) and flicker frequency ( F ) for response of the sunfish Lepomis ( Enneacanthus gloriosus ) to flicker is systematically displaced toward lower intensities by raising the temperature. The rod and cone portions of the curve are affected in a similar way, so that (until maximum F is approached) the shift is a nearly constant fraction of I m for a given change of temperature. These relationships are precisely similar to those found in the larvae of the dragonfly Anax . The modifications of the variability functions are also completely analogous. The effects found are consistent with the view that response to flicker is basically a matter of discrimination between effect of flashes of light and their after effects,—a form of intensity discrimination. They are not consistent with the stationary state formulation of the shape of the flicker curve. An examination of the relationships between the cone portion and the rod portion of the curves for the sunfish suggests a basis for their separation, and provides an explanation for certain "anomalous" features of human flicker curves. It is pointed out how tests of this matter will be made.

Curves relating flicker frequency ( F ) to mean critical illumination ( I m ) for threshold response to flickered light, with equal durations of light and no light intervals, and relating illumination ( I ) to mean critical flicker frequency ( F m ) for the same response, have been obtained from homogeneous data based upon the reactions of dragonfly larvae ( Anax junius ). These curves exhibit the properties already described in the case of the fish Lepomis . The curve for F m lies above the curve of I m by an amount which, as a function of I , can be predicted from a knowledge either of the variation of I m or of F m . The law of the observable connection between F and I is properly expressed as a band, not as a simple curve. The variation of I m (and of F m ) is not due to "experimental error," but is an expression of the variable character of the organism's capacity to exhibit the reaction which is the basis of the measurements. As in other series of measurements, P.E. I is a rectilinear function of I m ; P.E. F passes through a maximum as F (or I ) increases. The form of P.E. F as a function of I can be predicted from the measurements of P.E. I . It is pointed out that the equations which have been proposed for the interpretation of curves of critical flicker frequency as a function of intensity, based upon the balance of light adaptation and dark adaptation, have in fact the character of "population curves;" and that their contained constants do not have the properties requisite for the consistent application of the view that the shape of the F - I curve is governed by the steady state condition of adaptation. These curves can, however, be understood as resulting from the achievement of a certain level of difference between the average effect of a light flash and its average after effect during the dark interval.

The curve of mean critical flicker frequency as a function of illumination has been determined for the reaction of the sunfish Lepomis to flicker. It exhibits expected quantitative disagreements with the curve of mean critical illumination as a function of flicker frequency in the same organism. The form of the dependence of the variation of critical frequency of flicker upon illumination can be predicted from a knowledge of the way in which variation of critical illumination depends upon flicker frequency. It is pointed out that these findings have an important bearing upon the interpretation of the data of intensity discrimination.

The sun-fish Lepomis responds to a moving system of stripes by a motion of its body. By changing the velocity of motion of the stripe system different flicker frequencies can be produced and thus the relation of flicker frequency to critical intensity of illumination can be studied. Threshold illumination varies with flicker frequency in such a way that with increasing flicker frequency the intensity of illumination must be increased to produce a threshold response in the fish. The curve of critical illumination as a function of frequency is made up of two distinct parts. For an intensity range below 0.04 millilambert and flicker frequencies below 10 per second, the rods are in function. For higher intensities and flicker frequencies above 10, the cones come into play. The maximum frequency of flicker which can be perceived by the fish's eye is slightly above 50 per second. The flicker curve for the eye of Lepomis can easily be compared with that for the human eye. The extent of the curve for the fish is greater at low illuminations, the fish being capable of distinguishing flicker at illuminations lower than can the human eye. The transition of rod vision to cone vision occurs for the fish and for the human eye at the same intensity and flicker frequency. The maximum frequency of flicker which can be perceived is for both about the same.

Bees which are held in a fixed position so that only head movements can be made, respond to a moving stripe system in their visual field by a characteristic motion of the antennae. This reflex can be used to measure the bee's state of photic adaptation. A curve describing the course of dark adaptation is obtained, which shows that the sensitivity of the light adapted bee's eye increases rapidly during the first few minutes in darkness, then more slowly until it reaches a maximum level after 25 to 30 minutes. The total increase in sensitivity is about 1000 fold. The adaptive range of the human eye is about 10 times greater than for the bee's eye. The range covered by the bee's eye corresponds closely to the adapting range which is covered by the rods of the human eye.

1. For the phototropic reaction of bees, the stimulating effects of two illuminated fields differing in intensity and area become equal when the product of area and intensity is the same for both fields. 2. The effect of two areas differing in size and flicker frequency is the same for the bee, when the product of area and flicker frequency is equal for both fields. 3. If two patterns of the same character but varying in size and coarseness are presented to bees for free choice, a 1:1 ratio of choices is obtained when both patterns stimulate equal numbers of retinal elements alternately by transition from one state of excitation to another.

By presenting to bees two illuminated fields, equal in brightness, of which one is flickering and the other stationary we find that on account of the bee's positive phototropic response equal numbers of bees travel to both fields. We thus can assume that Talbot's law is valid for the eye of the bee.