E. Newton Harvey

Interest in the light of animals from a physical standpoint has centred around questions of quality, efficiency and intensity, but in only one group of luminous animals, the beetles, have accurate measurements of these characteristics been made. This is due in part to the abundance of these forms and their appeal to human interest and in part because they are among the brightest of luminous organisms. Weak lights are not only difficult to measure but, when dispersed to form spectra, give bands so faint that their limits are very difficult to see and more so to photograph. Very few organisms produce light visible to the fully light-adapted eye. Although their light may seem quite bright to the dark-adapted eye, the dark-adapted eye is a poor judge of the quality, i.e., the color of a light. This is because of the Purkinje phenomenon, a change in the region of maximum sensibility of the retina with change in intensity of the light. For an equal energy spectrum, to the normal, completely light-adapted eye, yellow-green light of wave-length, ? = .565µ, appears the brightest, but when the light is made fainter the maximum shifts first to the green and then to the blue. The dark-adapted eye can see green or blue better than yellow and for this reason weak lights will appear more green or blue than stronger ones of the same energy distribution. Also two weak lights of the same spectral composition may appear different in color if they differ much in intensity. This is illustrated in Fig. 6.

Fig. 6.—Visibility curves for three illuminations showing the shift in region of maximum visibility, or Purkinje phenomenon (after Nutting).

The shift in sensibility of the eye occurs in illuminations of between 0.5 and 50 metre-candles and represents a change from central cone vision (high intensities) to peripheral rod vision (low intensities). The fovea centralis lacks rods and this part of the eye becomes practically color blind at very low intensities of light. Below 0.5 and above 50 metre-candles visibility varies but little with change in intensity. It is clearly necessary then to distinguish between the physical objective phenomenon of light and the physiological subjective sensation of light.

It is a fact that different luminous animals produce light of quite different colors as judged by our eye. A range of spectral tints has been described which extends from red to violet but "yellowish," "greenish" and "bluish" tints are commonest. Indeed one or two animals possess several luminous organs emitting lights of different colors. This is true in a South American firefly, Phengodes, whose lights are red and greenish yellow, and in the deep sea squid, Thaumatolampas diadema, which produces lights of three colors, two shades of blue and red. The red light in the case of the squid appears to be due to a red color screen formed by the chromatophores, but in Phengodes no screen is present.

TABLE 4
Wave-lengths of Fraunhofer Lines and Prominent Lines in Line Spectra

FRAUNHOFER LINES

Line	Color	Wave-lengths (µµ = µ/1000)	Source
A	Red	759.4 (band)	Oxygen in atmosphere.
a	Red	718.5 (band)	Water vapor atmosphere.
B	Red	686.7	Oxygen vapor atmosphere.
C	Red	656.3	Hydrogen in sun.
D1 D2	Yellow	589.6, 589.0	Sodium in sun.
E	Green	527.0	Calcium in sun.
b1 b2 b4	Green	518.4, 517.3, 516.8	Magnesium in sun.
F	Blue	486.1	Hydrogen in sun.
G	Violet	430.8	Calcium in sun.
H K	Violet	396.9, 393.4	Calcium in sun.

BUNSEN FLAME LINES

Source	Color	Wave-lengths (µµ = µ/1000)
Potassium	Red	769.9, 766.5 (double)
Lithium	Red	670.8
Sodium	Yellow	589.6, 589.0 (double)
Thallium	Green	535.1
Magnesium	Green	518.4
Strontium	Blue	460.7

PLÜCKER TUBE LINES

Source	Color	Wave-lengths (µµ = µ/1000)
Mercury	Yellow	579.0, 576.9
	Green	546.1
	Blue	491.6, 435.8
	Violet	407.8, 404.7
Hydrogen	Red	656.3
	Blue	486.1, 434.1
Helium	Red	728.2, 706.5, 667.8
	Yellow	587.6
	Green	504.8, 501.6, 492.2
	Blue	471.3, 447.2
	Violet	438.8, 402.6, 388.8

As we have seen, difference in color of the light does not necessarily indicate difference in spectral composition because of the Purkinje effect. However, examination of the spectrum of various luminous forms has very clearly indicated that the different colors are really due to light rays of different wave-length and are not the result of any subjective phenomena. To facilitate comparison, spectral lines and colors are given in Table 4. The first adequate observations on the spectra of luminous animals were made by Pasteur (1864), who studied Pyrophorus and found a continuous spectrum unbroken by light or dark bands. Lankester (1868) discovered a similar continuous spectrum in ChÆtopterus insignis and placed its limits from line 5 to 10 on Sorby's Scale (about ? = 0.55µ to ? = 0.44µ). Young (1870) first recorded the limits of the firefly spectrum as a little above C (? = .6563µ) to F (? = .4861µ). Since then a number of luminous forms have been examined and all are found to give short continuous spectra (not crossed by light or dark bands or lines) lying in different color regions. Thus, Conroy (1882) examined the glowworm (Lampyris noctiluca) light and observed a band extending from ? = 0.518µ to ? = 0.656µ. Dubois (1886) states that the spectrum of Pyrophorus noctilucus, the West Indian "Cucullo," extends from slightly further than the Fraunhofer B line to the F line, while Langley and Very (1890), working on the same form, placed the limits at ? = 0.468µ to ? = 0.640µ. It consists, then, of a broad band chiefly in the green and yellow. But, "would the light not extend farther were it bright enough to be seen?... if the light of the insect were as bright as that of the sun would it not extend equally far on either side of the spectrum?" "It is impossible to increase the intrinsic brilliancy by any optical device, but if it be impossible to make the light of the insect as bright as that of the sun, it is on the other hand quite possible to make the light of the sun no brighter than that of the insect ..." Langley and Very investigated this question, forming a solar spectrum from sunlight of the same intensity as that of Pyrophorus and a Pyrophorus spectrum together in the same field of the spectroscope. The latter was very much shorter than the solar spectrum, showing that its length was not due to weakness of the red and blue rays but to their absence. Later Ives and Coblentz (1910) photographed the spectrum of a firefly (Photinus pyralis), together with that of a carbon glow lamp, on plates sensitive to all wave-lengths of visible rays under conditions which would have recorded all visible radiations given off. They found the spectrum to extend only from ? = 0.51µ to ? = 0.67µ (Fig. 7). Another species of firefly (Photuris pennsylvanica) was found by Coblentz (1912) to give a spectrum extending from ? = 0.51µ to ? = 0.59µ (Fig. 8). The Photinus light extends much further into the red and it is easy to distinguish between Photinus and Photuris in nature, merely by the reddish tint of the light of the former. These photographic records show conclusively that the color of the light of luminous animals is not a subjective phenomenon due to the Purkinje effect and the low intensity of the light, but is real, an actual difference in spectral composition of the light emitted. Neither is it due, at least in the fireflies examined, to the existence of color screens which absorb certain rays, allowing only those of a definite color to pass. The spectra of forms thus far investigated are reproduced in Fig. 9 and recorded in Table 5. It will be noted that they vary considerably in position but are all of the same type. The spectrum of Cypridina hilgendorfii is the longest thus far investigated (? = .610µ to ? = .415µ), extending well into the blue, and the light of this form is very blue in appearance.

---

Fig. 7.—Spectra of carbon glow lamp, A, firefly (Photinus pyralis); B, and helium vacuum tube, C (after Ives and Coblentz).

Table 5.—Limits of Spectra of Various Luminous Organisms

As first shown by Dubois (1886) for Pyrophorus, and confirmed by myself for Cypridina, the light is not polarized in any way. I may add that the Cypridina light like any other light may be polarized by passing through a Nicol prism.

Several writers [Dubois (1914 book)], Fischer (1888), Molisch (1904 book) have noticed that the light of luminous bacteria changes in color if grown on different culture media. Light which is "silver white" on dead fish becomes "greenish" on salt-peptone-gelatin media and more yellow on salt-poor media. Peron (1804) and Panceri (1872) describe the light of Pyrosoma as yellow to greenish after death of the animal and reddish on stimulation; then fading out through orange, yellow, greenish and azure blue. Polimanti (1911) describes the normal light of Pyrosoma as greenish, and states that as the animals die, or if they are kept at temperatures above the optimum, the light becomes more red. McDermott (1911, b) noticed that the light of fireflies placed in liquid air became decidedly reddish just before going out and on rewarming the first light to appear was reddish followed by the proper shade at higher temperatures. I have frequently observed a more reddish color from luminous tissues of the firefly upon the addition of coagulants such as alcohol, and have noted that the light of Cypridina becomes weaker and more yellow at both low (0°) and high (50°) temperatures. The meaning of these color changes will be discussed in Chapter VII.

The efficiency of any light may be defined in several different ways: (1) By the percentage of visible wave-lengths in the total amount of radiation emitted, i.e., visible radiation divided by total (heat, visible, actinic) radiation; (2) by considering, in addition to visible radiation ÷ total radiation, the sensibility of the eye to different wave-lengths, visible radiation × visual sensibility ÷ total radiation. Visible radiation × visual sensibility is spoken of as luminosity; (3) by the amount of light (expressed in candles) produced in relation to a given expenditure of energy or in relation to the cost of the energy expended. Thus, of the radiation emitted from an incandescent electric lamp only a small per cent. is light, the rest being heat and actinic rays. It is therefore very far from being 100 per cent. efficient. If there were no infra-red or ultra-violet in the radiation from an incandescent lamp its efficiency would be 100 per cent. if we disregarded visual sensibility. But if we take into account the fact that the eye is most sensitive to yellow green, a source of light, even though emitting only visible radiation, would not be 100 per cent. efficient unless its maximum of emission corresponded also with the maximum of visual sensibility. We shall return to this question in a later paragraph. Looking at the question from the standpoint of energy consumption, the carbon incandescent lamp gives one mean spherical candle for 4.83 watts (watt = 10⁷ ergs per sec.), while the tungsten lamp gives one mean spherical candle for 1.6 watts, about one-third the energy, and the latter is consequently more efficient.

As we know practically nothing of the energy transformations occurring during the process of light production in organisms, all statements regarding the efficiency of their light are based on relations between the visible radiation and total radiation. This involves a measurement of rays in the infra-red region (heat rays) and ultra-violet region (actinic rays) as well as the light rays proper, and any other radiant energy produced. While all spectroscopic investigations show that the spectrum of luminous animals never extends to the limits of the visible spectrum in either the red or violet, it is possible that bands occur in the infra-red or ultra-violet, and special methods must be employed to detect these. Radiations of all kinds, if converted into heat on striking the blackened surface of a thermopile, bolometer, or radiometer can be measured by changes in temperature and the relative amounts of energy represented be compared in a common unit, the calorie. By proper screening, all rays except the visible light rays can be cut off from the measuring instrument and the amounts of energy represented in light and in total radiation thus be determined.

Dubois (1886) first studied this problem in Pyrophorus by the use of a thermopile and galvanometer and found a small amount of radiation from the luminous region in excess of that from a non-luminous region. It amounted to a galvanometer deflection of 0.95° and was increased 0.3° during the flash of the insect on electrical stimulation. This increase of 0.3° is possibly due to heat produced on muscular contraction. In any case the amount of heat radiated in comparison with that of the candle is very small indeed. A more careful study has been made by Langley and Very (1890) with the bolometer. They point out first of all that the total radiation from the most powerful luminous organ (the abdominal one) of Pyrophorus which affected their bolometer slightly, would, in the same time (10 seconds), be sufficient to raise the temperature of an ordinary mercurial thermometer having a bulb 1 cm. in diameter by rather less than 2.3 × 10^-6° C. We may thus gain some idea of the magnitude of the measurements to be made. The radiation from Pyrophorus which affected their bolometer was shown to be due merely to the "body heat"[2] of the insect, and it is largely cut off by a plate of glass which is opaque to all wave-lengths of 3µ or more. These waves are given off by bodies at temperatures below 50° C. and belong "to quite another spectral region to that in which the invisible heat associated with light mainly appears." Langley and Very then compared the radiation from a non-luminous bunsen flame and the Pyrophorus light, interposing a plate of glass in each case to cut off the waves longer than 3µ, and found several hundred times more radiation in the case of the bunsen burner but, nevertheless, perceptible radiation from Pyrophorus. The former consisted of radiant heat shorter than ? = 3µ and extending up to the visible light rays (? = 0.7µ since the bunsen flame emitted no light). The very slight effect of the Pyrophorus radiation must be due to wave-lengths between ? = 3µ and ? = 0.468µ, the limit of the Pyrophorus spectrum in the blue. Langley and Very assumed it to be due entirely to the band of visible light, ? = 0.640µ to ? = 0.468µ, and assumed that no invisible heat rays were produced. All of the energy of Pyrophorus light would therefore lie in the visible region and its efficiency (light rays ÷ heat + light + actinic rays) would be 100 per cent. Later, Langley (1902) reinvestigated the radiation of Pyrophorus and could detect no heating whatever with the bolometer. "A portion of the flame of a standard sperm candle, equal in area to the bright part of the insects, gave under the same circumstances, a bolometric effect of such magnitude that had the heat of the insect been 1/80,000 as great as that from the candle, it would certainly have been recognized." Coblentz (1912) also, using a vacuum thermopile of Pt and Bi, was unable to detect any infra-red radiation from Photinus pyralis, but found that the temperature of this firefly is slightly lower than the air. These temperature measurements will be discussed in a later chapter.

The assumption of Langley and Very that the small amount of Pyrophorus radiation passing glass is all light has been called into question by Ives (1910), who points out that Langley and Very failed to use a screen which would cut off either the visible rays or the invisible rays between 3µ and 0.7µ. They really left the question open as to whether the effect of Pyrophorus light on their bolometer was due to the visible band of rays or to this plus another band in the infra-red. "The firefly's actual efficiency as a light source is dependent to a large degree on the radiation being confined to the visible region. If there should be found infra-red of quantity comparable to the visible, the firefly, while still a very efficient source would not be, as usually supposed, the example of an ideally efficient light produced by nature."

Ives investigated the question further by the phosphor-photographic method. "In brief it consists of this: Phosphorescence, which is excited in various substances by exposure to short waves (blue, violet or ultra-violet), is destroyed by exposure to longer waves (orange, red, infra-red). Thus, a surface of Balmain's paint or of Sidot blende, excited to phosphorescence and then exposed in a spectrograph, will have areas of reduced brightness wherever long-wave energy has fallen upon it. If this surface is then laid on a photographic plate for a short period, a permanent record is obtained on the plate after development." Preliminary tests showed that the method was applicable in the case of weak light such as the firefly spectrum and also if the light is intermittent like the firefly. With Sidot blend (ZnS) the extinguishing action extends from ? = 0.6µ to ? = 1.5µ. A sheet of deep ruby glass, which cut off all the visible rays of the firefly but allowed infra-red to pass, was placed between the firefly light and a surface of phosphorescent Sidot blend which was exposed to the firefly flashes for three and a half hours. No extinction of phosphorescence occurred, while without the ruby glass, extinction, due to the orange rays of the visible firefly light was noticeable in 20 minutes. There is thus no infra-red of an intensity at all comparable to the visible as far as ? = 1.5µ, the lower limit of the phosphor-photographic method. Coblentz (1912) had examined the transparency of the dry chitinous integument of various fireflies (Fig. 10) in the infra-red and reports it to be fairly transparent down to ? = 2.8µ, opaque between ? = 2.8µ and ? = 3.8µ, transparent again to ? = 6µ, and opaque beyond that. The infra-red could, then, if it were emitted, largely pass through the integument which is similar in absorption properties to complex carbohydrates. Transparency of the integument to the ultra-violet was not studied.

Although photographs of the spectrum of firefly (Photinus) light show that it extends only to the beginning of the blue, Forsyth (1910) reports ultra-violet radiation in luminous bacteria. He exposed a plate for 48 hours to the spectrum of bacterial light dispersed by a quartz prism and got a continuous band from ? = 0.50µ (the lower limit of sensitivity of the plate) to ? = 0.35µ. However, McDermott (1911 d) was unable to observe fluorescence of p-amino-ortho-sulpho-benzoic acid, which responds to the ultra-violet light. Molisch (1904, book) photographed bacterial and fungus light through glass and through a piece of quartz and found no difference in density on the plate. As the exposure was brief, to avoid saturation, and as the ultra-violet, which passes quartz but not glass, has a much greater action on the plate than visible light, we must conclude that ultra-violet is absent. Ives (1910) investigated the spectrum of Photinus pyralis, using a quartz spectroscope, and found no evidence of ultra-violet radiation, at least as far as ? = 0.216µ.

It will thus be seen that the radiation from the firefly has been very carefully studied and that no waves are given off from ? = 1.5µ to ? = 0.216µ with the exception of the short band (? = 0.67µ to ? = 0.51µ) in the visible, and it is highly probable that no radiation is given off with wave-lengths longer than ? = 1.5µ. The firefly light remains, then, 100 per cent. efficient, differing from all our artificial sources of light, the best of which does not approach this value. As Langley and Very express it in the title to their paper, it is "the cheapest form of light," not cheapest in the sense of that we can reproduce it commercially at less cost than other lights, but cheaper in the sense that it is the most economical in the energy radiated. This energy is all light and no heat. "Cold light" has actually been developed by the firefly and concerning which "we know of nothing to prevent our successfully imitating."

I have already pointed out that we may also consider the efficiency of a light in relation to the sensibility of our own eye. That is, we take into account not only the energy distribution in the spectrum of the light but also the fact that different wave-lengths of an equal energy spectrum affect our eye very differently. As the normal light-adapted eye is most sensitive to yellow green of ? = 0.565µ, monochromatic light of this wave-length will appear much brighter than monochromatic light of any other wave-length with the same energy. Monochromatic light of ? = 0.565µ will then be the theoretically most efficient possible, when we consider the energy radiated in relation to the sensitivity of our eye. This is the usual method of determining the luminous efficiency of artificial lights and is obtained from a knowledge of the radiated energy and the visual sensibility. Reduced luminous efficiency = light (radiated energy × visual sensibility) or luminosity ÷ total radiated energy.

The spectral energy curve for the firefly has been worked out by Ives and Coblentz (1910), using a photographic method in which the intensities of different wave-lengths of the firefly (Photinus pyralis) light is compared with that of a carbon glow lamp by measuring the amount of photochemical change produced on panchromatic photographic plates. Fig. 11 gives the energy curves of various fireflies and the carbon glow lamp in the same spectral region. The visual sensibility curve used by Ives and Coblentz is that of Nutting (1908, 1911), based on Konig's data. It is reproduced in Fig. 6. The latest visibility curve is that of Hyde, Forsyth and Cady (1918), reproduced in Fig. 12. It is based on observations of twenty-nine individuals. As individuals vary considerably in their sensibility to different wave-lengths, the visibility curve represents an average, but it is the only standard we have with which to evaluate the energy we call light. Color-blind individuals would have a visibility curve very different from normal individuals. Composite curves showing the luminous efficiency of the 4-watt carbon glow lamp and the firefly, both in relation to visibility, are given in Figs. 13 and 14, respectively. In these figures the luminous efficiency is the shaded area ÷ total area, 0.43 per cent. for the carbon glow lamp and 99.5 per cent. for the firefly, "these numbers representing the relative amounts of light (measured on a photometer) for equal amounts of radiated energy—a striking illustration of the wastefulness of artificial methods of light production. From the specific consumption of the tungsten lamp (1.6 watts per spherical candle) and the mercury arc (.55 watts per spherical candle) we obtained by comparison with the carbon filament that their luminous efficiencies are 1.3 and 3.8 per cent. The most efficient artificial illuminant therefore has about 4 per cent. of the luminous efficiency of the firefly." This is calculated to be .02 watts per candle. More recent determinations (Coblentz, 1912), using a new sensibility curve of Nutting's (1911) for a partially light-adapted eye, give the reduced luminous efficiency as 87 per cent. for Photinus pyralis, 80 per cent. for Photinus consanguineus and 92 per cent. for Photuris pennsylvanica.

The luminous efficiencies of various forms of artificial illuminants have been calculated by Ives (1915) and are given together with that of the firefly in Table 6. Fig. 15 gives spectral energy curves for various illuminants reduced to 100 at ? = .590µ, luminosity curves for the Hefner lamp and blue sky, and a visibility curve worked out by Coblentz and Emerson (1917) from observations on 130 individuals.

Table 6
Luminous Efficiencies of Various Illuminants

The firefly light by the above method of calculating efficiency is not 100 per cent. efficient because its maximum (? = 0.567µ) does not correspond with the maximum sensibility of the eye (? = 0.565µ), but taking into consideration also other effects of color, the firefly light would be a still more inefficient and trying one for artificial illumination, as all objects would appear a nearly uniform green hue. Indeed the distortion would be even greater than with the mercury arc, whose objectionable green hue is so well known. "We may say, therefore, that the firefly has carried the striving for efficiency too far to be acceptable to human use; it has produced the most efficient light known, as far as amount of light for expenditure of energy is concerned, but has produced it at the (inevitable) expense of range of color. The most efficient light for human use, taking into account both color and energy-light relationships, would be a light similar to the firefly light containing no radiation beyond the visible spectrum, but differing from it by being white." (Ives, 1910.) Although the spectral energy curve for Cypridina light has not been worked out, it will be noted that the Cypridina spectrum is much longer than that of the firefly, more nearly approaching the spectrum of an incandescent solid giving white light. It approaches, but does not attain the ideal.

Although Muraoka (1896) and Singh and Maulik (1911) have described radiations coming from fireflies which would pass opaque objects and affect a photographic plate, and Dubois reports the same from bacteria, the existence of such radiation has been denied by Suchsland (1898), Schurig (1901) and Molisch (1904 book). The experiments of Molisch on luminous bacteria are of greatest interest, for they are very carefully controlled and show without a doubt that black paper or Zn, Al, or Cu sheet will allow no rays from these organisms to pass that will affect a photographic plate, even after several days' exposure. The visible light of luminous bacteria will affect the plate after one second exposure. Moreover, Molisch has pointed out the errors of those who claim to have found penetrating radiation in luminous forms. It seems that certain kinds of cardboard, especially yellow varieties, or wood, will give off vapors that affect the photographic plate. The action is especially marked with damp cardboard at a temperature of 25°-35° C., and Dubois and Muraoka must have used such cardboard to cover their plates. A piece of old dry section of beech or oak trunk, placed on a photographic plate for 15 hours in a totally dark place, will register a beautiful picture of the annual rings of growth, medullary rays, junction of bark and wood, etc. Russell (1897) had previously found that many bodies, both metals and substances of organic origin (gums, wood, paper, etc.), placed in contact with photographic plates, would affect them, and concluded that vapors and not rays were the active agents. As a dry piece of wood has a very definite smell, there is something given off which can affect our nose and there is no reason why it should not change, by purely chemical action, the photographic plate. This action of wood on the plate is prevented by interposing a sheet of glass. Frankland (1898) has described similar vapors coming from colonies of Bacillus proteus vulgaris and B. coli communis which affect a photographic plate laid directly over the colonies in an open petri dish. There is no effect if the glass cover of the petri dish is between plate and bacteria. There is, then, no specific emission of X-rays or similar penetrating radiation from luminous tissues which will affect the photographic plate through opaque screens.

A similar conclusion is reached if we attack the problem in another way. X-rays and radium rays (Becquerel rays) cause fluorescence of ZnS, barium platinocyanide, willemite (Zn₂SiO₄), and calcium tungstate. Coblentz (1912) showed that the firefly will cause no fluorescence of a barium platinocyanide screen and I have been unable to detect fluorescence of zinc sulphide, barium platinocyanide, zinc silicate (willemite) or calcium tungstate shielded from Cypridina light by black paper, although the light of this organism is quite bright enough to cause phosphorescence of zinc sulphide without the black paper. The samples of the above four substances all showed fluorescence in presence of radium rays, but only the ZnS phosphoresces after exposure to light rays, although the willemite was phosphorescent after exposure to the ultra-violet.

While photometry at low intensities is a difficult procedure at best, if the light varies in intensity or is a flash, accurate measurements become well-nigh impossible. The figures given for intensity of animal luminescence must, therefore, be accepted with a realization of the difficulties of measurement. By candle is meant the international candle, unless otherwise specified, equal to 1.11 Hefner candles (H. K.) 0.1 pentane lamp and 0.104 carcel units. It is a measure of intensity.

Amount of light, or light flux, measured in lumens, is that emitted in a unit solid angle (area/r²) by a point source of one candle-power. One candle-power emits 4p lumens. The latest figure for the mechanical equivalent of light at ? = .566 is .0015 watt (Hyde, Forsyth and Cady, 1919), i.e., 1 lumen = .0015 watt. One watt is 10⁷ ergs (one joule) per second.

The illumination (of a surface) is that given by one candle at one metre, the candle metre (C.M.) or lux. The surface then receives one lumen per square metre. A metre kerze (M.K.) is the illumination given by one Hefner candle at one metre distance.

The brightness of a surface is measured in lamberts or millilamberts. A lambert is "the brightness of a perfectly diffusing surface radiating or reflecting one lumen per square cm." A millilambert is 1/1000 lambert. For further definitions the reader is referred to the reports of the committee on nomenclature of the Illuminating Engineering Society.

Dubois (1886) states that one of the prothoracic organs of Pyrophorus noctilucus has a light intensity of 1/150 Phoenix candle of eight to the pound (probably about equivalent to 1/150 candle) and that 37 or 38 beetles (each using all three light organs) would produce light equivalent to one Phoenix candle. Langley (1890) found that to the eye the prothoracic organ of Pyrophorus noctilucus gave one-eighth as much light as an equal area of a candle and the actual candle-power of the insect was 1/1600 candle. It may be remarked in passing how widely divergent these observations are.

For the flash of the firefly (Photinus pyralis) Coblentz (1912) found variation from 1/50 to 1/400 candle, the predominating values being around 1/400 candle. A continuous steady glow is sometimes obtained from this insect and it proved to be of the order of 1/50,000 candle.

Steady sources of light can be more easily measured and we have two records of the light intensity from luminous organisms with continuous light. One of these is a fish, Photoblepharon palpebratus, with a large luminous organ under the eye, of flattened oval shape, 11 × 5 mm., which glows continuously without change of intensity. The organ can be darkened by a screen similar to an eyelid which pulls up over it. Steche (1909) reports the intensity to be .0024 M.K.[3]

Luminous bacteria probably glow with less intensity than any other organism. The light from a single organism cannot be seen but that from a colony is visible to the dark-adapted eye. Even so we must remember that the eye is an exceedingly delicate instrument which can detect very small energy changes. The "minimum radiation visually perceptible" has been calculated by Reeves (1917) to be in the neighborhood of 18 × 10^-10 ergs per second and the light from a small colony of luminous bacteria represents little more radiation than this.

Lode (1904, 1908), by a modified grease spot photometer method, ascertained that the light of his brightest bacterial colony of Vibrio rumple had an intensity of 7.85 × 10^-10 H.K. per sq. mm. or 0.785 H.K. per 1000 sq. metres (=0.562 German-normal candles per 1000 sq. metres). In round numbers this is about one German-normal candle per 2000 sq. metres, or two to three times this area for the light from an ordinary stearin candle. Lode calculated that the dome of St. Peter's at Rome, if covered with bacteria, would give little more light than a common stearin candle. An ordinary room of 50 sq. metres wall and ceiling area would give out only 0.039 German-normal candle. It does not seem likely that luminous bacteria will ever come into vogue for illuminating purposes. Friedberger and Doepner (1907) by a photographic method, not entirely free from error, found that one square millimetre of lighting surface of a bouillon culture of photobacteria gave 6.8 × 10^-9 German-normal candles, about ten times Lode's value. Even at this rate commercial lighting by luminous bacteria does not appear a promising field for investors.

To sum up, we may say that light from animal sources is in no way different from light of ordinary sources, except in intensity and spectral extent. It is all visible light, containing no infra-red or ultra-violet radiation or rays which are capable of penetrating opaque objects. It is not polarized as produced, but may be polarized by passing through a Nichol prism. Like ordinary light, animal light will also cause fluorescence and phosphorescence of substances, affect a photographic plate, cause marked heliotropism of plant seedlings (Nadson, 1903) and stimulate the formation of chlorophyll (Issatschenko, 1903, 1907). Because of the weakness of bacterial light, etiolated seedlings do not become green to the eye (Molisch, 1912 book), but a small amount of chlorophyll is formed which can be recognized by the spectroscope because of its absorption bands.