MAY FLIES have long come in for considerable attention from those who seek to tempt the trout or to tempt the pocketbooks of trout-fishermen by making artificial flies. Occasionally this purely utilitarian interest leads a man into the realm of amateur entomology. Such an individual is Richard Salmon, an artist whose off-hours are devoted to angling and to research into the life cycle, color variation and related phenomena of the May flies inhabiting the eastern U. S.

Salmon's interest in the May fly began when he turned to it for guidance and inspiration in constructing artificial flies for his fishing. He soon found himself deep in the study of aquatic entomology. Through the years he has read everything he could find on the subject and has spent much time on fishing trips digging under rocks and exploring riffles for varieties of May flies.

He has recorded his findings not only in writing but in the form of exquisitely fashioned artificial flies and in detailed paintings and in drawings such as those reproduced with this article. Last year he exhibited some of his work at the American Museum of Natural History in New York. The display comprised 100 outstanding examples of the fly-tyer's art. It included materials and standard colors used in making the imitations, a series of beautifully executed drawings of the natural insects and a description of their entomological background-all of which represented many years of field research along the streams and lakes of the East.

No one knows who invented the artificial fly. Perhaps some ancient cave dweller lashed animal hair or feathers to a bone hook and, squatting on the banks of a Paleolithic river, became the first to cast a home-made fly into a stream. One of the earliest writings on the subject was by the Roman teacher Aelian, who recorded that the Macedonians used flies to fish in the river "Astraeus" between Messalonica and Borea. In the 15th century Dame Juliana Berners, Prioress of Sopwell Nunnery and an authority on hunting, hawking and "Fysshynge with an Angle," described "flyes with which ye shall angle to ye trought" which closely resembled the flies used by trout anglers today. Made of "donne woll, wyngis of the pertryche" and other materials common to the modern fly-maker, they simulated the May flies that dance over the English chalk streams in the spring and summer twilight.

May flies play an important part in the ecology of trout. Ordinarily during their season they constitute nearly 80 per cent of the trout's diet. They are most in evidence during a very brief period when they emerge from the nymphal state and swarm over the waters to mate and die. Like the Ephemerides of Greek mythology, May flies are said to live only one day, from this comes the insect order's technical name, Ephemerida.

Richard Salmon's studies have gone much further than the casual observations of the hobbyist; they encompass the whole life-cycle of the May fly from egg to adult. He collects the flies and nymphs "by hand," rather than with conventional tools of the entomologist such as the butterfly net.

"Catching the flies is not difficult," he explains. "The floating insects come to hand easily; those in flight travel slowly enough so that a vigorous wave of my hand carries them off their course and into the water where they may be picked up. In the nymphal underwater stage they are more difficult to catch. I take a small window screen to the stream, place it below the area I intend to search, and stir up the bottom. The current carries nymphs into the screen

"I have a series of small clear-glass, wide-mouthed bottles in which I keep a solution of formalin and water: one part 50 per cent formalin to 20 parts water. To this I add a couple of drops of glycerin, which keeps the insects pliable. I use screw-top bottles and find it wise to keep them brim-full, because splashing in a partly filled container breaks their delicate bodies.

"I have never succeeded in keeping specimens in full color for any length of time. Their colors fade. When drawing and water-coloring, I find work should be done as soon as possible after collecting.

"It is advisable to carry a magnifying glass for identification purposes. The flies are easily distinguishable from all other insects by their remarkably large forewings and small hind wings, very short antennae, two or three long curving tail-filaments and entirely atrophied mouth parts. In this stage they are known colloquially as 'Drake,' 'Day Fly,' 'Sailor,' 'Lake Fly,' 'Shad Fly' and by other names. They are recognizable by nearly everyone who has found himself in the vicinity of streams where the insect breeds in astronomical numbers. In some towns along the St. Lawrence River the swarms of May flies are sometimes so dense that they stop traffic. They cover sidewalks with a slippery mess which pedestrians are unable to negotiate. May flies apparently mistake the black, shiny surfaces of macadam roads for stream surfaces. They often dip down and deposit eggs on road-tops as if they were rippling streams."

Although folklore gives the insect a lifetime of only one day, it really lives a great deal longer, most of the time under water.

"The life cycle is interesting and unique," Salmon reports. "Eggs are usually deposited in fresh water with a high oxygen content. A few species choose brackish water. In a short time the eggs hatch into minute nymphs, which are mainly herbivorous, living on diatoms and other minute water plants. They swim, crawl and burrow, according to their kind. Most species may be found under stones in or near swift water. The nymphs are quite active, with strong legs, distinguished by a single terminal claw.

"Physical idiosyncracies for the identification of ephemerid nymphs are few. In the main they have two shapes. Some are flat, their compressed bodies streamlined for the purpose of hugging the bottom. Their shape deflects the rapid current Flat nymphs are usually the crawlers and the burrowers. Others are round-bodied. These are usually the swimmers. Some travel at remarkable speed and are almost impossible to catch by hand.

"During the insect's early growth there are several molts. The potential May fly grows faster than its skin, which quickly sloughs off to expose a new and larger shuck of similar pattern. The nymph's life-span usually lasts from one to three years. It may have as many as 20 molts. In one of the larger species (Hexaga Variabilis) the body of the pupa or nymph is an inch long in the final stage, with tail filaments extending half an inch more.

"As the nymph passes through the adolescent stage, it develops humps on its shoulders, where later the wings of the adult will emerge. When it is ready to transform, its wing-pads swell, giving it a hunchbacked appearance. It becomes very active, finally swimming or floating to the surface of the water. Then the skin of its back splits open and the winged insect flies out. In some kinds, such as the one known to fishermen as the 'Pale Evening Dun,' the transformation is so rapid that the fly seems to explode from the nymphal shuck. Time and time again I have observed the nymph rising to the surface; the next instant, without any perceptible struggle, it is in the air, flying rather awkwardly on its untried wings.

"The rather large March brown, one of the Stenonema, is quite the opposite, and equally interesting to watch. The nymph swims to the surface in a zigzag series of spurts. It floats, wiggling in snapping convulsions, trying to crack through its shuck. I have followed the nymph downstream for a hundred yards watching its struggles. Finally the fly pulls itself from the burst skin and, apparently exhausted, sits upon its own skin while its wings dry. Then it tries to fly, and failure follows miserable failure until somehow it succeeds in becoming airborne. Because of its long period on the surface, trout find this fly easy to catch. The fish seem to go into a frenzy as they wait for the fly to emerge.

"Fishermen describe the transformation of aquatic flies as a 'hatch,' and a big 'hatch' is an exciting spectacle indeed. One moment the water will present a glassy surface without so much as a ripple; then suddenly the whole area becomes pock-marked-a frenzy of 'hatching' flies. Almost instantly the agitation spreads. The whole stream comes to life-charged with excitement. Swallows, flycatchers and other birds swoop over the water and gorge themselves on the sudden feast. And every trout in the stream, it seems, joins the party. At this point I cease temporarily to be an amateur entomologist, and my metamorphosis into a fisherman becomes as astonishing as that of the Ephemeridae!

"In the stage after transformation the May fly is known as a subimago, or, in the parlance of the fisherman, a 'dun.' The subimago may be distinguished from the true adult, or imago, by its direction and pattern of flight and by its cloudy, opaque color.

"Upon emergence it makes a beeline shoreward, where it conceals itself under a leaf or in some other inconspicuous place. It clings to its place of sanctuary, resting and calming down after its exciting experience. Then a curious phenomenon takes place, one not found in the life cycle of any other insect. There is a second molt. The skin of the subimago splits and the true adult fly emerges. At times I have seen this second transformation take place in mid-air.

"The life of the imago is extremely short. The fly begins to starve as soon as it leaves the nymphal shuck. It cannot eat because its mouth-parts are atrophied. The life span is usually longer, however, than the fabled 24 hours some species live as long as four days if the air is dry and the atmosphere quiet. The l9th-century British entomologist John Curtis is said to have kept an adult May fly alive indoors for three weeks

"The imago-called 'spinner' by trout fishermen-is a far brighter and more beautiful insect than the subimago. Its wings are glassy and finely veined. For a long time the order Ephemerida was classified under the Neuroptera, the nerve-winged flies, because of the finely laced veins of their wings.

"The males are easily distinguished from the females. In a female May fly the legs are too delicate to support the weight of the body, whereas in a male only the hind legs are weak.

"May flies usually remain quiet during the heat of the day and fly in the cool of the morning or evening. They are attracted by artificial light and swarm around street lights or the headlights of automobiles. Sometimes the swarms are as thick as a dense snow-cloud.

"Most May flies couple during flight, the male clinging undermost. Egg-laying follows immediately. The male, spent falls or flies off weakly to die, usually in a matter of minutes. Some May flies drop their eggs from mid-air. In most cases, however, the female proceeds in undulating flight, dipping down to the surface of the water to wash off loose clusters of eggs from her ovipositor. She rises to dip again and again. It is this undulating flight pattern that is known as the 'dance of the May fly.'

"During the egg-laying dance the insects continually work upstream. It has been suggested that this is a means of preserving the species. In swift water the eggs drift some distance before settling to the bottom. Thus if the flies merely dropped their eggs over the area where they themselves were hatched, each generation would hatch farther downstream than its predecessor, and conceivably the whole species might eventually be lost in the sea.

"In many once fine trout streams the numbers of May flies have sharply decreased, and in all too many they no longer exist at all. This lamentable decrease results from a combination of causes: pollution, flash floods that leave silt deposits, and the destruction of forests, which leaves streams unshaded and lets them fall in summer to low levels so that the water becomes too warm for the flies.

"The loss of May flies is of considerable concern to fishermen. During the latter part of the l9th century transplantations of English 'Drakes' were made successfully to America. Nymphs of the insect were planted in New York's Navesink River, where they propagated and continued to flourish. Some day, when industry stops dumping untreated factory-wastes into the waters, when the farmer and lumberman have at last adopted sound soil and timber conservation methods, we may see May flies clouding the air above streams in their former numbers-and let us hope there will be trout splashing for them as they rise from the foam-flecked pools and eddies as they did in the good old days."

TWO IMPORTANT recent books illustrate why diffraction gratings are increasingly used instead of prisms in research spectrographs. In Practical Spectroscopy George R. Harrison, Richard C. Lord and John R. Loofbourow of M.I.T. point out that a grating costing $800 will do the work of a quartz prism costing $10,000. In Fundamentals of Physical Optics Francis A. Jenkins and Harvey E. White of the University of California show that a quartz prism with resolving power comparable to that of a six-inch grating (of 15,000 grooves per inch) would have to be 80 inches wide-an impracticable size. It would be half as large as a pup tent, an object of similar shape.

The grating on the June cover, with 5 1/8-inch ruled width, rates in spectroscopy as a six-inch grating because the measure used is the diagonal, the diameter of the light beam.

Diffraction gratings make spectra by an optical mechanism altogether different from that of prisms. According to classical theory, in a prism electrons vibrating in the atoms of the glass fall into resonance with the light waves and retard them. The resonance is greater for the shorter waves; hence the violet is retarded more than the red. This is what causes the differential bending that is responsible for dispersion of the colors in the light. In a grating the light reflected from a single groove is diffracted (deflected) over a wide angle. The light thus diffracted from many identical grooves then interferes, producing spectra.

With these wide differences in optical mechanism it would be remarkable if the spectra of prisms and gratings were alike. The illustration at the bottom of the next page shows how greatly they differ. The most striking difference is the fact that, while a prism spectrum is single, grating spectra are multiple. The lower part of the drawing shows the first four of many recurring pairs of spectra made by a grating. At the center is a white reflected image of the light source, the slit of the spectrograph. The same mechanism which puts light into a number of orders puts some of it into this zero order, but largely its light is reflected from the submicroscopically torn and burred edges and bottoms of the grating grooves, which cannot be ruled with geometrical perfection.

On both sides of the central image, in corresponding left- and right-hand pairs, are numbered "orders" of spectra. (R G and V stand for red, green and violet. Each pair is wider and fainter than the last. After the first order they overlap in apparent confusion. To a spectroscopist this is not as bad as it appears. From daily familiarity he comes to know his way about. He selects for use a single order on a single side, ignoring the others. Usually he removes the others with a filter. The chief advantage in the grating spectra is the fact that the resolving power is multiplied theoretically in proportion to the order selected. Practical reasons reduce this bonanza, so that the third order may no more than double the resolving power of the first, and with the fourth the gains begin to dwindle markedly.

There are other fundamental differences between prism and grating spectra. The upper part of the drawing shows a prism spectrum. The prism bends red rays the least, violet the most, whereas a grating does the opposite, the red rays falling farthest from the central image and the violet nearest.

Finally, a grating disperses the wave lengths almost uniformly. This is shown in the small drawing above, where the wavelengths are given in Angstrom units, from 4000 A. in the violet to 7000 and beyond in the red. In the grating spectrum (top) the spacing is even; in the prism spectrum (bottom) the wavelengths are increasingly crowded together toward the red. In the "rational" grating spectrum identification and measurement of wavelengths is much easier. This is what originally interested Henry A. Rowland in gratings.

Until recent years a serious drawback of the grating spectrograph was the fact that the total light was divided between the central image and many orders of spectra, so that the individual spectra were illuminated only faintly. A way was found, however, to make them brighter. Rowland had noted that gratings accidentally ruled with lopsided (not V-shaped) grooves sometimes gave a brighter spectrum in some single color than a prism did. Later the experimental physicist R. W. Wood described similar phenomena. The British physicist Lord Rayleigh investigated the theory of these anomalous gratings and proved that this special distribution of light was connected with the cross-sectional shape of the grooves, as suspected. Wood finally discovered that if he shaped the ruling diamond and set it at such an angle that the light reflected from one side of the groove coincided in direction with the diffracted light, most of the light could be thrown into the part of the spectrum where maximum illumination was desired. Today this is called the "blaze."

A chisel-edge diamond ground to shape and tilted for blazing a grating is shown in a drawing (left). Its keel is 1/16-inch long; the grooves shown are far out of scale in size. After the diamond has ruled about 10 miles of grooves (say three six-inch gratings), its leading corner will be rounded by wear. At the Johns Hopkins University the ruling engine technician Wilbur Perry expertly regrinds and repolishes worn diamonds on a lapidary's lap, removing about 1/10,000 inch of thickness in the process. The keel of the one that I examined under 700 diameters' magnification was too sharp to be seen; it was "visible" only as a diffraction line. Compared with its smooth edge, a razor's edge is a crosscut saw.

In 1916, when John A. Anderson went from Johns Hopkins to the Mount Wilson Observatory as an astrophysicist and ruling-engine consultant, he developed the "boat-bottom" diamond shown below the chisel-edge. Its curved edge is obtained by the intersection of two cones, one shallow, one deep. Only the shaded portion is used, set in metal. The curved edge presses the metal aside by plastic flow, also burnishing the groove. This form of diamond works wonderfully well on plane gratings, but poorly on the concave gratings that Rowland invented; in place of an added lens, for focusing the light.

A grating may be blazed not only for a single order on one side of the central image but even for a definite part of that order. To a physicist this advance little known outside the world of physics, is outstandingly significant and almost as important as more obvious advances such as larger gratings. Setting the diamond and checking the control of the blaze calls for patient fussing. Several gratings may be ruled before the correct angle is hit upon. David Richardson, in charge of the ruling engines at the Bausch & Lomb Optical Company, has been a leader in the art of blazing gratings, "doing a bang-up job on the control of the grooves," as a friendly competitor remarks. He accomplishes much of this control with a microinterferometer instead of a microscope, thus "seeing" the shape of the grooves in terms of interference fringes.

While it is easy to see the grooves with a high-power microscope, it is difficult to interpret their shape. Harold D. Babcock explains: "Every microscopist is familiar with the uncertainties of interpreting the appearance of a periodic structure when the limit of resolution is approached. Spurious details, delicately responsive to the slightest variation in the focal adjustment or in the illumination, are particularly troublesome in field of parallel equidistant lines."

The pioneers Wood, Anderson and Babcock blazed gratings in hard, brittle intractable speculum metal. Two decades ago John Strong, also Robley C Williams, developed practical method of depositing aluminum on glass. This metal is comparatively soft and tractable and has high reflectivity that lasts a long time. Its arrival greatly stimulated blazing, and thus the advance of the diffraction grating toward supremacy over the prism. Today a grating can concentrate more of the incident light in the blaze than a prism, with its large losses by internal absorption and surface reflection can put into its single spectrum. Perry's gratings average 60 per cent of the light in one order, frequently reach 80 percent and have reached 95 per cent.

John Strong remarks that it is just a wrong to discourage anyone from trying to build a ruling engine as to over encourage him. But the victim should be fully aware of the difficulties. Rowland' assistant and alter ego, the physicist J. S. Ames, listed Rowland's special combination of qualifications for ruling-engine work: 1) the scientist's grasp of fundamental principles, 2) the engineer's understanding of practical mechanics, 8) mathematical aptitude, 4) manual dexterity. Understanding of mechanics and high manual dexterity are not uncommon among amateurs with toolmaking skill. It is suspected, however, that some who have tackled the ruling engine and judiciously kept it a secret, or tried to keep it a secret from this department's elaborate spy network, have underestimated the importance of science and mathematics. Learning with surprise that most of the physicist operators o ruling engines do not themselves run the engines but make only occasional calls like a physician, while the technicians stay with the engines like a nurse, some have concluded that the "nurse" does the work while the "doctor" gets the credit. They do not realize how essential are the doctor's guiding judgment, diagnosis and treatment when things go wrong, as they usually do.

Rowland (who, by the way, pronounced the first syllable of his name to rhyme with doe and not with cow) acknowledged his technician Theodore Schneider but once in print, although Schneider made the screws and most of the working parts of the Rowland engines and superintended the ruling of every grating that left Johns Hopkins for a quarter-century. Today ruling-engine physicists are more careful to give public credit to their technicians in scientific periodicals. Horace W. Babcock acknowledges contributions by his technicians C. Jacomini and E. D. Prall, and Strong praises the work of Wilbur Perry, John F. McClellan, Dave Broadhead, Howard Head, Robert Schreitz and William Koenig.

Some of the earlier ruling-engine operators were secretive. Today they are mutual friends who pool most of their findings in "family sewing bees." If any seem less communicative to new aspirants, the reason may be only that they are weary of hearing the-same old proposals for engine designs advanced as new, often not without self-confidence. But their reticence may also be an act of mercy. After publication of the article on ruling engines in the June issue an enthusiastic mechanic informed me that he had made up his mind to borrow money from a bank to build a ruling engine and turn out 14-inch gratings "within six months." At the other extreme, the same article caused another fine mechanic, a doctor of science, to give up his long-smoldering ambition to build a ruling engine.

Scale drawings and a description of the Rowland engine were published in the Physical Papers of H. A. Rowland. This book is now rare, but photostatic copies of pages 691-697, on the ruling engine, and separate figures 1 to 5 may be purchased from the Library of Congress, Washington, D.C. No scale drawings of other engines have been published. The classic article of J. A. Anderson, "The Manufacture and Testing of Diffraction Gratings," occupies pages t30-41 of Volume 4 of Richard Glazebrook's Dictionary of Applied Physics. Anderson's article on "Periodic Errors in Ruling Machines," in Journal of the Optical Society ot America, July, 1922, pages 434-442, also is basic to the understanding of the subject. Most of the literature on the ruling engine is cited by George R. Harrison in the same journal, June, 1949.

The Society for Amateur Scientists (SAS) is a nonprofit research and educational organization dedicated to helping people enrich their lives by following their passion to take part in scientific adventures of all kinds.