(page 285)

Witch flounder Glyptocephalus cynoglossus (Linnaeus) 1758


[Jordan and Evermann, 1896-1900, p. 2657.]

Witch flounder (Glyptocephalus cynoglossus)

Figure 146.—Witch flounder (Glyptocephalus cynoglossus). From Goode. Drawing by H. L. Todd.


The witch or "gray sole" as it is now named in the United States fishery statistics, is right-handed (viscera on the right hand as the fish lies) and small-mouthed like the winter flounder, the smooth flounder and the yellowtail. But there is little danger of confusing it with any of these for its fin rays are much more numerous, its body narrower relatively, its head much smaller, and the open mucous pits on the blind side of its head large and conspicuous. It is two and one-half to three times as long as it is broad (deep, in reality), elliptical in outline, very thin but with its head occupying only about one-fifth of the total body length, and it has a very small mouth. The dorsal (left-hand) profile of its head is convex. It has 100 to 115 dorsal fin rays and 87 to 100 anal rays, and the anal fin is preceded by a short, sharp spine pointing forward, which is a prolongation of the post-abdominal bone. The two long fins are of about uniform width throughout most of their lengths, except that they narrow gradually toward head and tail. The pectoral fins and the ventral fins are alike on the two sides, or nearly so, while the caudal fin is much smaller, relatively, than that of the yellowtail, of the winter flounder, or of the smooth flounder, though similarly rounded in rear outline.

The lateral line is straight, as a rule, but it is somewhat arched abreast the pectoral fin in some specimens. The teeth are small, incisorlike, and in a single series. There are about 12 open mucous pits or depressions on the blind side of the head, and less obvious ones on the eyed side also. The whole body and head (except for the tip of the [page 286] snout and the lower jaw) are scaly, but the scales are smooth to the touch, which make the witch as slippery to hold as a female smooth flounder (p. 284).


By all accounts (and the fish we have seen are in line with this) the witch is less variable in color than most of the flatfishes. Most of them are brownish or russet gray on the eyed side, either uniform or with darker transverse bars, with the vertical fins of the general body hue, tinted or tinged with violet, and either plain or spotted. The pectoral fin membrane on the eyed side is dusky or even black, a feature distinctive of this particular flatfish. The lower (blind) side is white, and more or less dotted with minute dark points. An occasional fish is colored on the under side as well as on the upper side; one of this sort, 19 inches long, was landed at the Boston Fish Pier early in March 1931.


The maximum length is about 25 inches, and fish of 23 or 24 inches, weighing about 4 pounds, are not uncommon. But the general run of those caught are only about 12 to 20 inches long.


The witch flounder is rather a deepwater fish, seldom caught shoaler than 10 or 15 fathoms once it has taken to bottom, though taken occasionally close inshore (see footnote 93, p. 288). Off the American coast the best catches are made between about 60 fathoms and about 150 fathoms. Thus the Albatross III caught an average of about 57 witch flounders per trawl haul at 100 to 150 fathoms on the southwestern part of Georges Bank in mid-May 1950. But an average of only about one fish per haul between 31 fathoms and 80 fathoms. And they have been trawled widespread down the continental slope as deep as 858 fathoms off southern Nova Scotia; to 732 fathoms off Marthas Vineyard; to 788 fathoms off Chesapeake Bay; and to 602 fathoms off North Carolina.[84]

In Swedish waters, according to Melander,[85] the best catches are made between 80 and 140 fathoms.

They are caught most abundantly on fine muddy sand, on clay, or even on mud. They are said to frequent hard reefs in Scandinavian waters, but this does not seem to be the case in the Gulf of Maine, though they are common there on the smooth ground between rocky patches.

When the witch has once taken to the bottom it seems to be even more stationary in our gulf than some other flounders, for it is caught the year round, with no evidence that it moves in or offshore with the change of the seasons. In Swedish waters, however, it is said to work up into shoaler water in autumn, and deeper again in late winter and spring.[86]

It occurs in the Gulf of Maine in temperatures ranging from about 35°-38° F. (late winter and early spring), to 45°-48° (late summer and early autumn), according to precise locality and depth. In the Gulf of St. Lawrence it occurs in the icy [page 287] cold waters (30°-32°) on the banks as well as in the higher temperatures (40°-42°) of the deep channels.[87] Apparently it is never found in any numbers in water warmer than 50°, but we hesitate to propose high temperature as the factor barring it from shoal water because there is no evidence that it works inshore in our gulf in winter when this bar would not operate.


It feeds on invertebrates, like other small-mouthed flatfishes; European experience points to small crustaceans, starfish, small mollusks, and worms, as its chief diet.[88] It is not known to eat fish and does not take a bait often.

Breeding as it does through a long season, over many degrees of latitude, and in both sides of the Atlantic, the witch spawns in temperatures ranging from close to the freezing point of salt water up to 48°-50° F. (p. 288). And experiments, added to captures of eggs naturally spawned, and of newly hatched larvae, have shown that incubation proceeds normally in water at least as cold as 45°-46° F., and as warm as 50°-55° F.

The eggs are buoyant, spherical, transparent, with narrow perivitelline space (the perivitelline space is broad in the eggs of the dab or Canadian plaice, which overlap them in dimensions), without oil globule, and 1.07 to 1.25 mm. in diameter. As noted (pp. 288 and 203), there is danger of confusing newly spawned witch eggs with those of the cod and haddock, for they overlap these in size and in season. But identity is easily recognizable after a few days' incubation, for black pigment is to be seen in the gadoid eggs soon after the embryo is visible as such, but does not appear in the witch flounder eggs until after hatching.

Incubation occupies 7 to 8 days at temperatures varying from 46° to 49° F., and the newly hatched larvae are about 4.9 mm. long, with a larger yolk sac than those of our other flatfishes. The yellow and black pigment becomes aggregated into five transverse bands on body, yolk (now much reduced in size), and fin folds within a few days after hatching, when the larva is 5 to 6 mm. long. One of these bands is at the region of the pectoral fin, one at the vent, and three of them on the trunk rearward from the vent. The yolk is entirely absorbed in about 10 days after hatching, the caudal rays have begun to appear at a length of 15 mm., the rays of the vertical fins are well advanced at 21 mm. and they are complete in their final number at about 30 mm. The eyes are still symmetrical, or nearly so, up to this stage. But the left eye has moved to the dorsal surface of the head in larvae of about 40 mm. And the migration of the eye is complete at a length of 40 to 50 mm., when the young fish takes to the bottom.

Witch Flounder (Glyptocephalus cynoglossus). Egg. Larva, Smallest bottom stage.

Figure 147.—Egg (European) After Cunningham.

Figure 148.—Larva (European), 10 days old, 5.6 mm. After Holt.

Figure 149.—Larva (European), 16 mm. After Kyle.

Figure 150.—Smallest bottom stage (European), 42 mm. After Petersen.

Witch Flounder (Glyptocephalus cynoglossus)

The witch is perhaps the most easily recognizable of Gulf of Maine flatfishes throughout its [page 288] larval stage. The transverse pigment bars are diagnostic prior to the appearance of the caudal rays, while the curiously concave ventral profile of the throat region with the comparatively long slender trunk are equally so, thereafter. And the great number of dorsal and anal fin rays, coupled with the small mouth, make identification easy after the fins are formed. The witch also grows to a larger size before it completes its metamorphosis than does any other of the right-handed, small-mouthed flatfishes that are found in the Gulf of Maine.

Measurements of the young (American as well as European), suggest that the free-drifting stage may last as long as 4 to 6 months for the witch, which is much longer than for any of our other flatfishes.

Fry of 2½ to 4¼ inches, and of 3½ to 45/8 inches, such as we have trawled in July and August, respectively, probably are in their second summer, their sizes depending on how early in the season they were hatched the year before. The subsequent rate of growth has not been traced for American fish. If Molander's[89] estimate for European fish is correct, the size group centering at 6¾ to 8 inches that was prominent in our August catches of 1936 were in their third summer.[90] And subsequent growth is very slow.

General range—

Moderately deep water in both sides of the North Atlantic. Its European range is from northern Norway and Iceland south to the west coast of France. In American waters its free-drifting larvae are reported from as far north as the Strait of Belle Isle, around the coasts of Newfoundland, and over the Grand Banks region in general.[91] The adult is known from the Gulf of St. Lawrence; the south coast of Newfoundland; the southern part of the Grand Banks; in Cabot Strait; along outer Nova Scotia and the Scotian Banks; throughout the Gulf of Maine; and thence westward and southward along the continental shelf and slope as far as the offing of northern Virginia (lat. 37° 50' N.) in moderate depths, to the offing of Cape Hatteras in deep water.[92]

Occurrence in the Gulf of Maine—

The distribution of this flatfish in our Gulf is governed by the fact that it is a fish of at least moderately deep water, seldom caught as shoal as 10 fathoms.[93] In fact, its very existence remained unsuspected by Massachusetts fishermen until 1877, when the United States Fish Commission caught numbers of them while trawling in the deeper parts of Massachusetts Bay. Since that time it has been reported (or we have trawled it, or both) from St. Mary Bay on the Scotian side of the Gulf; in the Bay of Fundy and its tributaries (where Huntsman describes it as taken very generally below 15 fathoms, if not in any great numbers); at Eastport; off Mount Desert, where we have trawled it as shallow as 10 fathoms; near Monhegan Island; off Seguin Island; off Cape Porpoise; near the Isles of Shoals (where Welsh saw a few taken from the gill nets set in about 25 fathoms in April 1913); in the deep trough to the westward of Jeffreys Ledge; in Ipswich Bay; near Gloucester; off Boston Harbor; at various localities in the deeper parts of Massachusetts Bay; and in both branches of the deep trough of the Gulf west and east down to a depth of 140 fathoms; in the deep channel, between Browns Bank and Georges Bank, and on the slope to the southeast.

Trawlers bring them in regularly from Browns Bank, also from Georges, where Welsh found them widespread, and from Nantucket Shoals.

This is enough to show that the witch is to be expected anywhere in our Gulf where the water is deeper than 15 to 20 fathoms, if the bottom is suitable.

The largest catches are made on the so-called South Channel grounds which include the slopes that lead down from the offing of Cape Cod on the one side and from Georges Bank on the other, into the southwestern part of the basin; farther north off eastern Massachusetts; and off western Maine. And the published statistics suggest that gray soles are about as plentiful as the American dabs are on the various grounds where the trawlers work regularly.

Reported landings of gray soles by New England vessels in 1947 were as follows for the several statistical areas:[94] Browns Bank, 44,000 pounds; off western Nova Scotia, 2,000 pounds; off eastern Maine, 17,000 pounds; off central [page 289] Maine, 12,000 pounds; off western Maine, 630,000 pounds; small grounds in west central part of Gulf, 77,000 pounds; off eastern Massachusetts, 582,000 pounds; South Channel grounds, east and west, 629,000 pounds; other parts of Georges Bank, 94,000 pounds; Nantucket Shoals region, 16,000 pounds.[95]

More precise evidence as to their local numbers on suitable bottoms in the appropriate depths is that as much as 500 pounds have been taken in a 15- to 20-minute haul with a small beam trawl in Massachusetts Bay, and that we caught 48 of them in Ipswich Bay in 22 fathoms, in a short haul with an 8-foot beam trawl on July 16, 1912. We also saw 519 of them, 10 to 22 inches long, trawled on the southwestern part of Georges Banks by the Eugene H, in 41 hauls at 26 to 65 fathoms in late June 1951, and learned that this dragger caught 9,000 pounds on the northeastern edge of Georges Bank, in 85 to 95 fathoms, October 12-18, 1951.

Neither the witch flounder nor the American dab is as plentiful as the yellowtail on good flounder bottoms, or the flatfishes of the winter flounder group (blackbacks plus lemon soles). And Atlantis took only 156 witch to 279 dabs on soft bottom at 90 to 103 fathoms during experimental trawling in the deeper parts of the Gulf in August 1936.

Gray soles are at least moderately plentiful off southern New England. The Albatross III, for example, took 90 there in one trawl haul at 101 to 150 fathoms in mid-May 1950, a few as shoal as 31 to 40 fathoms. And a few thousand pounds are landed yearly in New York and in New Jersey ports.[96] But records of the witch from farther south than New Jersey are of an occasional fish only.

Reported landings suggest that gray soles are about as plentiful all along the Nova Scotian banks as they are in the Gulf of Maine region. In 1947, for example, New England vessels landed about 555,000 pounds of them from the various grounds from the eastern part of Browns Bank to Banquereau, about half of which came from the Horseshoe ground between Halifax and Sable Island.[97] And they seem to be moderately plentiful in the southern part of the Gulf of St. Lawrence, for Cox[98] wrote of many (large and small) as taken off the Cape Breton shore, and in Cabot Strait off Cape North. But no information is available as to their numbers elsewhere in the Gulf of St. Lawrence, or on the Grand Banks.

It seems that the witch does not breed successfully in the Bay of Fundy; at least its eggs have never been found there, nor have its larvae. But probably it does so in other parts of the Gulf in general, including the offshore Banks, though our only positive egg records for it have been off Penobscot Bay, and at the mouth of Massachusetts Bay. And there is no reason to doubt that the more northerly populations are equally self supporting, for the pelagic larvae have been taken at many localities on the more easterly of the Nova Scotian Banks; in the Gulf of St. Lawrence; over the Grand Banks; and along the south and cast coasts of Newfoundland, by the Canadian Fisheries Expedition of 1915, and during the cruises of the Newfoundland Fishery Research Commission more recently.[99] But there is no evidence that the witch spawns to any extent to the west of Cape Cod.

Captures of eggs, certainly of this species, in our tow nets in July and August, with larvae up to 20 to 23 mm. long as early as the first week of July, but others as small as 9 to 10 mm. as late as mid-October, show that the witch is a late spring and summer spawner in the Gulf of Maine as it is in European waters also, with the peak of production probably falling in July and August. Thus its spawning season overlaps that of the haddock. (p. 207).

Its eggs are shed in the Gulf of Maine in temperatures ranging from 39° to 41° F. at the beginning of the season, to 43° to 48° in midsummer. But (being buoyant) the temperature may be considerably higher at the level where their development takes place than deeper down where the spawning fish lie. In fact, it is doubtful if any eggs develop in our Gulf in water as cold as 42° to 43°. Neither is there any reason to suppose that witch eggs develop in water any colder than this in the Gulf of St. Lawrence, or off Newfoundland, for the surface stratum to which they rise after [page 290] they are shed is comparatively warm (upward of 45°) in these seas also, during the spawning season.

At the other extreme, our captures of eggs and of newly hatched larvae near the surface in July prove that the latter may be hatched in the Gulf in water at least as warm as 50° to 55°. But the upper limit to normal development cannot be stated from the evidence yet in hand, for with a temperature gradient as steep as it is over most of the Gulf of Maine in summer a difference of only a few fathoms in the depth at which the eggs or young larvae are suspended may mean a difference of several degrees of temperature.

One result of the protracted spawning season, combined with the long period occupied by larval development, is that witch larvae of various sizes are to be taken in tow nets throughout the summer and early autumn, as appears from the following table of our catches on the Grampus.

Date Number of larvae Length in millimeters
July 7, 1915 109 8 to 23.5
July 8, 1913 19 8.5 to 21.5
July 9, 1913 1 14
July 19, 1916 100+ 5 to 19
July 22, 1912 1 9.5
July 24, 1912 2 8.5 and 16.5
Aug. 5, 1913 27 5.5 to 12.5
Aug. 9, 1913 7 10 to 23
Aug. 14, 1912 1 18.5
Aug. 15, 1912 3 18.5 to 37.5
Aug. 24, 1912 6 10 to 18
Aug. 25, 1914 19 10 to 19
Aug. 26, 1913 2 8 and 14
Aug. 29, 1916 100+ 5 to 19
Aug. 31, 1912 20+ 9 to 16.5
Sept. 29, 1915 22 10 to 14
Oct. 18, 1915 1 9.5
Nov. 1, 1916 20+ 29.5 to 50

All of these catches, like those for other larval flatfishes, and for larval gadoids, have been concentrated in the southwestern part of the Gulf, which must be an important nursery for the witch also. And we may note in passing that the presence of young fry at all stages from immediately after their metamorphosis (that is, 4 to 6 months old) in the Bay of Fundy, where few or none are hatched, points to an immigration of the late larvae, or of the youngest fry, into the Bay, either just before they take to the bottom or soon after they have done so.


The witch was of no commercial importance in our Gulf a quarter of a century ago; few fishermen distinguished it from other flounders then and no record was kept of the catch. It is an excellent table fish; and the bases of its fins are provided with astonishingly large amounts of gelatinous fat for so thin a flounder, of the sort for which the European turbot is famed.

It is now in such demand that it brings about as high a price as either the yellowtail or the American dab. In 1947, for instance, the average price at Massachusetts ports was about 7 cents for gray sole, about 8 cents for yellowtail, about 7 cents for dab, about 9 to 12 cents, according to size, for flatfish of the winter-flounder type (blackbacks and lemon sole), about 17 cents for summer flounders (fluke), and about 25 cents for halibut. The Gulf yielded between 2 million and 2½ million pounds of gray sole both in 1946 and in 1947, corresponding to something like 1 to 1¼ million individual fish.

The otter trawl is the only gear now in use in our waters that is adapted to the capture of witch flounders on a commercial scale. They live too deep, and their mouths are too small for them to be of any concern to small-boat fishermen.

[84] Goode and Bean (Smithsonian Contr. Knowl., vol. 30, 1895, p. 433) give a long list of deep-water stations for the witch off southern New England.

[85] Pub. de Circonstance No. 85, Cons. Internat. Explor. Mer, 1925, p. 3.

[86] Melander, Pub. de Circonstance, Cons. Internat. Explor. Mer. No. 95, 1925, p. 3.

[87] According to Huntsman, Trans. Roy. Soc. Canada, Ser. 3, vol. 12, Sect. 4, 1918, p. 63.

[88] No witch flounder stomachs have been examined in the Gulf of Maine, so far as we know.

[89] Based on the structure of the otoliths; Pub. de Circonstance, No. 85, Cons. Perm. Internat. Explor. Mer. 1925, pp. 12-14.

[90] Bigelow and Schroeder, Biol. Bull., vol. 76, 1939, pp. 318-319.

[91] See Frost, Res. Bull. No. 4, Newfoundland Dept. Nat. Resources, 1938, Chart 6, for Newfoundland localities.

[92] Goode and Bean (Smithsonian Contrib. Knowl., vol. 30, 1895, p. 433) list it from lat. 34° 39' N., 603 fathoms.

[93] A stray specimen, picked up in a pound net at Eastport, Maine, many years ago, was reported by Gill (Proc. Acad. Nat. Sci., Philadelphia, 1873, p. 360) as a new species, Glyptocephalus acadianus.

[94] To the nearest 1,000 pounds.

[95] An additional 182,000 pounds were landed in Cape Cod fishing ports, source not stated.

[96] About 19,000 pounds in New York in 1947, about 28,000 pounds in New Jersey.

[97] Only a few thousand pounds are reported yearly from Nova Scotia, in the Canadian fishery statistics.

[98] Contrib. Canad. Biol. (1918-1920) 1921, p. 113.

[99] See Frost, Research Bull. 4. Newfoundland Dept. Nat. Resources, 1938, Chart 6, for Newfoundland localities.