ANNOTATED LIST OF THE SPECIES 149 appears to be the shelter of the rafts used as farmstead canoe landings, or floating docks, where much of the life of the farm takes place. Among these activities are the butchering and fish-drying for the household, and the offal and fish blood are disposed of through the chinks. Sr. Julio Battistini showed me the method of taking these fishes with the greatest ease and rapidity. Without pole, line, hook, or sinker, a simple piece of fish-flesh was dangled between the logs of the raft. The carnero attacked the bait with such pertinacity that from one to three could be pulled up and shaken off into a basin at a time.
Among the natives they have the reputation of being tropic to urine, and are said to enter the body of female bathers only, due to their large size, while other species attack males.
My first collecting trip, which was confined to the coastal and cordilleran regions, did not penetrate into the tropical habitat of the candirv. But everywhere I met people familiar with the stories about it, and it is probably unknown to but very few of the inhabitants of Peru.
107. HeMIcETOPSIS PLUMBEUS (Steindachner) Cetopsis plumbeus Steindachner, 1882 (1883), Denksch. KKK. Akad. Wiss. Wien, XLVI, 31, pl. vi, fig.3,Canelos,Ecuador,two,60and70mm.; Boulenger,1887,Proc.Zool.Soc.London,276,Canelos; EKigenmannandEKigenmann,1890,Oce.PapersCal.Acad.Sei.,I,320; Pearson, 1924, Ind. Univ. Studies, XI, no. 64, 16, Rio Bopi, Bolivia, ten, 74-140 mm.; Pearson, 1937a, Proc. Cal. Acad. Sci., (4), XXIII, 94, Pusoe and Tingo de Pauea, upper Maranon.
HemicetopsisplumbeusEigenmann,1910,Rept.PrincetonUniv.Exped.Patagonia,IIT,398. Known only from Canelos and the two widely separated habitats listed by Pearson.
Differing from the preceding in that there are two or three series of premaxillary teeth, that the anterior nares are less widely spaced than the posterior; gillopenings higher, reaching above first pectoral ray; inner ventral ray adnate to belly wall; pectorals shorter, not reaching the ventral fins. Genus 53: PSEUDOCETOPSIS Bleeker PseudocetopsisBleeker,1862,Versl.Akad.Amst.,XIV,403; HKigenmann, 1910, Rept. Princeton Univ. Exped. Patagonia, III, 398. Cetopsis(Pseudocetopsis)EigenmannandEigenmann,1890,Occ.PapersCal.Acad.Sci.,I,398- Type: Cetopsis gobioides Kner Both sides Ecuadorean Andes to Central Brazil Without conical or incisor-like teeth, or at most a single band of conical teeth onvomer.
108. PskUDOCETOPSIS VENTRALIS (Gill) Cetopsis ventralis Gill, 1870, Proc. Acad. Nat. Sci. Phila., XXII, 95, Napo or Maranon basin, Orton collection; [> [LIBRARY] 39 = > WEN =
150 FISHES OF WESTERN SOUTH AMERICA Eigenmann and Eigenmann, 1890, Oce. Papers Cal. Acad. Sci., I, 322. PseudocetopsisventralisEKigenmann,Rept.PrincetonUniv.Exped.Patagonia,ITT,398. Known only from Orton’s specimens at the Philadelphia Academy Museum. Subfamily: PYGIDIINAE Mostly or altogether free-living South American Nematognaths, having the dorsal fin over or behind the ventrals; adipose fin wanting; opercle and interopercle spinous; barbels reduced to a pair at each angle of the mouth, and otherwise irregular; frequently of elongated, almost eel-like form, but very diverse in form and habit; generally intrusive, by virtue of the spines and hooks which anchor them to the sides of an opening, as they elbow themselves among the rocks of rapids and rifles or burrow in sand or mud, or insinuate themselves into the cavities and gillchambers of other fishes.
Fishes of the highlands and scantily dispersed into adjacent lowlands, rarely down to near sea-level. Next to Orestias the dominant fishes of the high Andes. Genus 54: PARIOLIUS Cope Pariolius Cope, 1871 (1872), Proc. Acad. Nat. Sci. Phila., XXTIT, 289; HKigenmann and Eigenmann, 1889, Proc. Cal. Acad. Sci., (2), II, 50; EFigenmannandHigenmann,1890,Oce.PapersCal.Acad.Sci.,I,324; Gosline,1940,Copeia,no.2,78.
Type: Pariolius armillatus Cope Range that of the species fan) Dorsal fin in front of the ventrals; no nasal barbels; opercle and preopercle without spines; a single barbel on each maxillary and two pairs of mentals. Near Pygidium, but not conforming to the boundaries of the subfamily; probably out of order here, and nearer the main trunk of the Stegophilines. Gosline considers it a nearer ally of the Pimelodidae.
109. PARIOLIUS ARMILLATUS Cope Pariolius armillatus Cope, 1871 (1872), Proc. Acad. Nat. Sci. Phila., X XTIT, 289; EigenmannandEigenmann,1890,Occ.PapersCal.Acad.Sce1.,I,324; Eigenmann,1910,Rept.PrincetonUniv.Exped.Patagonia,IIT,398; Gosline,1940,Copeia,no.2,79.
Rio Ambyiacu With new material from R. Ambyiacu Gosline has redescribed this fish. Genus 55: PYGIDIUM Meyen Trichomycterus Valenciennes, 1833, in Humboldt, Rec. d’Obs. Zool. Anat., H, 348, non Thrichomycterus Humboldt.
Pygidium Meyen, 1835, Reise in Peru, I, 474, fuscum; Eigenmann and Eigenmann, 1890, Oce. Papers Cal. Acad. Sci., I, 325;
ANNOTATED LIST OF THE SPECIES 151 Kigenmann, 1910, Rept. Princeton Univ. Exped. Patagonia, III, 399; Eigenmann, 1918c, Mem. Carnegie Mus., V, 288.
Type: Pygidium fuscum Meyen Panama to Chile and Patagonia, inhabiting mountain lakes and streams, rarely lowland streams to near sea level, the species usually limited to a few closely connected drainage areas Of elongate body form, cylindrical to terete, caudal peduncle compressed and elevated; without scales or plates, two barbels at each angle of mouth, no mental, generally short; eye small and at or before middle of head, without free orbital border; interopercle with several series of spines; gill-membrane narrowly united with isthmus, and usually a narrow, free margin across it; mouth terminal, moderate; Jaws with two or more chisel-like teeth, sometimes conical; no labial or vomerine teeth; fins without spines, although sometimes filamentous; ventral fins at or beyond middle of body.
110. PyGipium ruscum Meyen PygidiumfuscumMeyen,1835,ReiseinPeru,I,475; Tschudi,1845,FaunaPeru.,21; EKigenmann, 1910, Rept. Princeton Univ. Exped. Patagonia, III, 399; Eigenmann, 1918c, Mem. Carnegie Mus., VII, 298.
Peru, in the High Andes Aside from Tschudi’s assurance that P. fuscum is distinct from his own dispar, there is no corroboration of the existence of this species with its unsatisfactory description. Not even the locality is known, and may not be in that portion of Peru covered by this report. Or it may have originated in that portion of the larger viceregal Peru now known as Bolivia.
111. Pyarprum virratum (Regan) Trichomycterus vittatus Regan, 1903, Ann. Mag. Nat. Hist., (7), XII, 623. PygidiumvittatumEigenmann,1910,Rept.PrincetonUniv.Exped.Patagonia,III,400; Eigenmann,1918c,Mem.CarnegieMus.,VIT,299.
Marcapata valley, southeastern Peru Only the types are known. Origin of dorsal before vertical from anus, and its distance from the caudal one and one half the distance from the snout; anal wholly caudal of the vertical from the last dorsal ray; distance of the base of the last anal ray from the caudal four and one half times in the total length; caudal truncate; dark spots on head and body, and a dark lateral stripe. 112. PyGiprum pisparR Tschudi Pygidium dispar Tschudi, in part, 1845, Fauna Peru., 22, pl. ili, upper figure, High Andes, eastern slope, 14,000 feet elevation;
152 FISHES OF WESTERN SOUTH AMERICA EigenmannandEigenmann,1890,Occ.PapersCal.Acad.Sci.,I,335; Pellegrin, 1904, Bull. Soe. Zool. France, X XIX, 91; Pellegrin,1907,Poiss.LacsHautsPlat.Amér.Sud,17; Eigenmann, 1910, Rept. Princeton Univ. Exped. Patagonia, III, 400; Evermann and Radeliffe, 1917, Bull. U. S. Nat. Mus., no. 95, 35: Eigenmann, 1918c, Mem. Carnegie Mus., VII, 299, pl. xlv, fig. 5. Trichomycterus dispar Giinther, 1864, Cat. Fish. Brit. Mus., V, 273; Cope, 1877, Proc. Amer. Phil. Soc., XVII, 46; Cope, 1878, Proce. Amer. Phil. Soc., XVII, 681, Urubamba, 10,000 feet elevation; Tinta, 11,400 feet.
Fic. 19. The Rio Pacaya of the Ucayali basin. Twin-motored canoe of large size whichwasusedonthisstream.Itswatersarebrown,andinfestedwithcaymans. Higher elevations both sides of the Andes and L. Titicacz Cope’s specimens may be P. rivulatum. Cope regards Giinther’s specimens as maculatum by reason of its smaller number of dorsal radii. With this we can agree except that Cope also appears to confuse dispar with rivulatum. Giinther’s characterization fits the latter species as well as dispar. He says also that it occurs in all parts of Lake Titicaca and its tributaries; eats small molluses, crustacea, and fishes; a 14-inch specimen contained one of six inches; brown, reticulated with many white lines, sometimes spotted like a trout; adult specimens of less than seven inches up to 16; the name maure given to the large ones, suche to the younger ones. Cope says the young have a more or less interrupted, lead-colored lateral band which breaks up with growth.
113. Pyaipium RIvuLATUM (Cuvier and Valenciennes) TrichomycterusrivulatusCuvierandValenciennes,1846,Hist.Nat.Poiss.,XVIII,495,Guasacona; Cope, 1877, Proce. Amer. Phil. Soc., XVII, 46, Lago Titicaca, one large specimen, too few dorsalraystobedispar;
ANNOTATED LIST OF THE SPECIES 153 Pellegrin, 1904, Bull. Soc. Zool. France, X XIX, 91; Pellegrin,1907,Poiss.LacsHautsPlat.Amér.Sud,17,LagoPoopé,Pazna. Pygidium rivulatum Eigenmann and Eigenmann, 1889, Proc. Cal. Acad. Sei., (2), I, 51, Cuzco, Moho,Puno,LakeTiticaca; Eigenmann and Bean, 1907, Proc. U.S. Nat. Mus., XX XI, 771; EKigenmann,1918,Mem.CarnegieMus.,VII,301; Pearson, 1937a, Proc. Cal. Acad. Sci., (4), X XIII, 94, Cajamarca. ?Trichomycterus incae Cuvier and Valenciennes, 1846, Hist. Nat. Poiss., XVIII, 496, Cuzco, Rio Guatanal.
TrichomycterusgracilisCuvierandValenciennes,1846,Hist.Nat.Poiss.,XVIII,497,RioAzangaro nearGuasacona;RioGuatanai,Cuzco;LagoCompucilanearCuzco; Cope, 1877, Proc. Amer. Phil. Soc., XVII, 681, Tinta. TrichomycterusbarbatulaCuvierandValenciennes,1846,Hist.Nat.Poiss.,XVIII,498,Guasacona; Pontezuela,nearCoroico,Bol.
TrichomycteruspentlandiCastelnau,1855,Anim.Amér.Sud,Poiss.,49,pl.xxiv,fig.1,“lakecommunicatingwiththeUcayali.”
TrichomycteruspictusCastelnau,1855,Anim.Amér.Sud,Poiss.,59,pl.xxiv,fig.2,LagoTiticaca. TrichomycterusdisparGarman(nonTschudi),1875,Bull.Mus.Comp.Zool.,III,275,LagoTiticaca; Pellegrin, 1904, Bull. Soe. Zool. France, XXIX, 91; Pellegrin, 1907, Poiss. Lacs Hauts Plat. Amér. Sud, 7, Lago Titicaca. Trichomycterus pardus Cope, 1874, Proc. Acad. Nat. Sei. Phila., XX VI, 132. Central plateaux of southern Peru and adjacent highlands 13833,1,98mm.,Tirapata,13000feet,purchasefromRosenberg. 13750, 2, 123 and 140 mm., Ollantaytambo, Rio Urubamba, E. Heller. 15063,many,30-82mm.,MediaLuna,RioUrubamba,Eigenmann,November,1918. 15064, many, , Calacoto, Rio Desaguadero, Bolivia, Allen, 1919. 15065, 3, up to 300 mm., Puno Market, L. Titicaca, Allen, December, 1918. 15066,14,42-150mm.,Ollantaytambo,R.Urubamba,Eigenmann,November,1918. 15068,74,106-156mm.,Saylla,Cuzcovalley,Eigenmann,November,1918. 15072,7,104-122mm.,RioUrubamba,Eigenmann,November,1918. 15073,many,40-133mm.,Angostura,Cuzcovalley,Eigenmann,November,1918. 15074, many, 30-125 mm., Saylla, Cuzco valley, Eigenmann, November, 1918. 16058, 2, about 400 mm., Pto. Acosta, L. Titieaca, Allen, December, 1918. 16059, , , Lago Titicaca, Puno, Allen, 1918.
17795,3,60-69mm.,Torontoy,R.Urubamba,Eigenmann,November,1918. 17796,many,386-178mm.,LagoUreos,Eigenmann,November,1918. 17797, many, 39-215 mm., Aguas Calientes, R. Vileanota, Allen, January 1919. 17798, many, 41-90 mm., Lago Verde, R. Vileanota, Allen, January, 1919. 17809,many,32-102mm.,lagoon,Huaypo,Eigenmann,November,1918. 17810, many, 31-95 mm., Poroy, Eigenmann, November, 1918. 17811, , R. Huatanay, Eigenmann, November, 1918. 17812,26,39-73mm.,Iscuchaca,R.Huatanay,Kigenmann,November,1918. 17813,many,52-96mm.,Angostura,R.Huatanay,Higenmann,November,1918. 17814,many,54-179mm.,Hermita,Oropesa,Higenmann,November,1918. 17815, 5, 37-58 mm., R. de Langui, Eigenmann, November, 1918. 17816,26,24-86mm.,R.Occopata,Eigenmann,November,1918. Of the specimens numbered 17810 and 17816 none has the protracted ray of the pectoral well-developed, suggesting atochae. They are also mostly small and not robust.
Generally known as the suche (elsewhere locally called bagre, in a generic sense; also boga and maur?).
Our largest specimens were taken near Moho on Lake Titicaca, at Puerto Acosta, by means of a home-made wicker lead-in trap resembling the lobster-pot,
154 FISHES OF WESTERN SOUTH AMERICA with a funnel beneath, and suspended from a balsa by means of a native grass rope. In general the largest and best-conditioned specimens came from lakes. Pearson (1924) collected P. rivulatum as far down into Bolivia as the fringe of the foothills at about 1500 feet elevation. His specimens show variations which are about constant within each locality. He finds none of the typical vermiculations of the upland specimens, but the spots are large or small, in horizontal lines, or irregular. His fluviatile specimens have larger barbels than those of lakes. His collections and those reported above show that diligent search has been made over a wide territory without bringing to light any specimens which we could identify as P. dispar, which may therefore fall within the wide range of variation of rivulatum, in case a few errors have been made in describing it. While not a rival in numbers or amount marketed in the Titicaca valley, it outranks its competitor Orestias in size and quality. The flesh is soft and there are few bones. Suggesting Ameturus of North America in form, the bottominhabiting habit is likewise similar.
114. Pyaiprum oroyAE Eigenmann and Eigenmann Pygidium oroyae Eigenmann and EFigenmann, 1889, Proc. Cal. Acad. Sci., (2), I, 51; EigenmannandEigenmann,1890,Occ.PapersCal.Acad.Sci.,I,334,8specimens,R.Oroya, Pochachara; Eigenmann, 1910, Rept. Princeton Univ. Exped. Patagonia, III, 399; Evermann and Radeliffe, 1917, Bull. U.S. Nat. Mus., no. 95, 35, pl. iv, fig. 2, Oroya; Eigenmann, 1918, Mem. Carnegie Mus., VII, 304, text fig. 10. Rio Mantaro, central Peru, having a very restricted range, not extending far from the valley of the small river which gives it the name, except that it has invaded the upper Perené.
17799, many hundreds, 20-206 mm., Rio Oroya and Rio Mantaro, Oroya, Eigenmann and Allen, September, 1918.
17800,manyhundreds,20-102mm.,Tarma,R.deTarma,Eigenmann,September,1918. — , Many, , Tarma, R. de Tarma, and springs, Allen, June, 1920. —, 18, 24-90 mm., Tarma, Lola E. Vance, about 1912.
17801,11,35-135mm.,Zigzag,branchR.Mantaro,Allen,October,1918. 17802,2,137-184mm.,creek,L.Chinchaycocha,Allen,October,1918. 17803, many, 36-110 mm., Tilarnioc, R. Mantaro, Eigenmann, September, 1918. 17804,2,94and115mm.,Pachachaca,Eigenmann,September,1918. 17805,many,22-82mm.,Huancayo,Eigenmann,September,1918.
17806,26,20-71mm.,Quebrada,Huanecayo,Kigenmann,September,1918. 17807,1,115mm.,Jauja,Eigenmann,September,1918.
17808,3,830-77mm.,Palea,Peru,Eigenmann,September,1918.
Originally described from eight specimens, the description amplified by Evermann and Radcliffe, 1917, from the examination of two. We have before us many hundreds reaching a maximum of 206 mm. compared with former maxima of 140 mm. and 130 mm., and collected from a wide range of localities. The dorsal fin overlaps the anal; ventrals much reduced; the caudal rounded in the younger, becoming subtruncate in the larger; pectoral short and rounded, first ray not prolonged; anal fin large; fins becoming thick and fleshy with age; lappet at angle of mouth triangular and bulbous.
ANNOTATED LIST OF THE SPECIES 155 Markings obscure in older specimens, becoming more olivaceous with age, while in rivulatum the spots become more pronounced in the larger. Higher up the Mantaro, in the slower, more sluggish streams and lakes of the pampas (13000 feet and above), such localities as Zigzag, Tilarnioc, and Junin, we find the specimens more elongate, the lappets not usually bulbous, and the color brown with no indication of an olivaceous tint.
Our first specimens were collected in the diminutive type stream, the Rio Oroya, where it rushes precipitously among rocks, boulders, and rubble for a halfmile to join the Rio Mantaro. The roots of the cubé plant were used here for the first time; the bruised pulp and milky extract were placed in the stream at the top of the canon, into water so turbulent that you would not expect to find any life. But ina short time the Pygidiwm began to boil out from among the rocks in enormous numbers, struggling for breath, and were easy victims to the dipnet. This was a surprise even to the bystanders.
115. PyarpiuM TACZANOWSKIL (Steindachner) Pygidium dispar Tschudi, part, 1845, Fauna Peru., 22, pl. iii, lower figure. TrichomycterustaczanowskiiSteindachner,1882,Denksch.KIX.Akad.Wiss.Wien,XLVI,22,pl.iv, figs. 1-1b, Rio de Huambo, Rio de Tortora.
Pygidium taczanowski Eigenmann and Eigenmann, 1889, Proc. Cal. Acad. Sei., (2), II, 52; Eigenmann and Eigenmann, 1890, Oce. Papers Cal. Acad. Sci., I, 338; Eigenmann, 1910, Rept. Princeton Univ. Exped. Patagonia, III, 400; Higenmann, 1918, Mem. Carnegie Mus., VII, 301, pl. xlvi, figs. 5-8; Pearson, 1937a, Proce. Cal. Acad. Sci., (4), XXIII, 93, Maranon. Central Peru northward in moderate elevations 17817,17,130-226mm.,RioHuallaga,Hudnuco,Allen,October,1918. 17818,8,45-69mm.,RioTingo,Hudnuco,Allen,October,1918.
The latter specimens all much smaller in this small tributary stream than in the nearby Huallaga into which it flows.
Steindachner was given the local name kutschin for this species. My informants usually applied the name bagre, meaning catfish, though others were calling it huascachallhua.
In many respects our Hudnuco specimens show a close correspondence to the original descriptions: D.9; A. 7; P. 9; width of head in its length 1.3; snout 2.1—-2.3; interorbital 3.3-near 4, as compared with the broader 3.0-3.33 of the original description; nasal barbels 1.3; lower barbels 1.6; width of mouth in head 2; eye in front of middle to middle of head. Our specimens unlike the types in that the origin of the dorsal is nearer the caudal than to the gill-opening in those of about 130 mm., and in our largest 1.4 nearer the caudal as compared with 1.22 nearer in the types. The ratio of the first pectoral ray to the length of the head about the same. ‘These measurements indicate greater variability in the young rather than a progressive advancement caudally of the dorsal fin with age.
156 FISHES OF WESTERN SOUTH AMERICA | Taczanowskil Taczanowsku Characters Rivulatum | Oroyae type from Rio Tingo Atochae Head in length |4.5-5.5* | 5.75-6.0 | 4.75-6.5* [4.65.3 | 4.76.3 Dorsal | 13* ae a0 i 8-9 fleshy i Le ie -_ \i011* | 7 73 6-7 fleshy Pectoral Vs fie Iso =—ti(iéidrt SQ Nasal barbell in | toend of eye iow i 25-1.4 | 1 3 mad 2 9-2 ..6 head Wvaeiitars: 2 - Be ee - none reaching a .21-1 5 1 re a 1 8-2. 0 barbel same opercle gill-opening Mental barbel a _ 1.6-2.0 | f 6t 2. 1 2.40 same Caudal founded il roundeae” - eaesente subtnincate . Pant Denk in feneen (arses . “5.75-8.0 7 6.3-8.4 6.0-7.0 8.3-9.8 - Dorsal to anal overlaps half overlaps some- barely reaches* barely reaches* overlaps — half way what way Accessory rays distinet* few and in- few and in- few, not dis- | Moderately _ numerous distinct distinct tinct numerous In the above brief tabular analysis, certain characters have attained their highest development in one or more species studied. P. rivulatwm has or shares the extreme position in six cases, oroyae four times, the taczanowskii types three cases, the taczanowskii specimens from the Tingo four times, and atochae eight times. Intermediate positions are held by rivulatwm three times, oroyae seven, taczanowskii types seven, taczanowskii from the Tingo eight times, and atochae only three times. Atochae shares no characters as listed with révulatum, but does so once with oroyae and twice with taczanowskii. Thus the characters of the above species seem to arrange themselves in a series, following the order in which they are given, rivulatum the most extreme in robustness and in number of finrays, etc., with atochae its opposite. (One extreme shown by * ; opposite extreme by italics). 116. Pyeiprum aTocHAE Allen, sp. nov.
Plate XIII; figs. 3-5 17819, many, type and paratypes, 38-105 mm., Rio de Atocha, Atocha, Bolivia, Allen, February, 1919.
Allied to Pygidium taczanowskii, oroyae, and rivulatum, in that order, differing most strikingly in its slender form and short barbels; middle of eye in advance of middle of head; body less compressed midway; caudal peduncle distinctly more elongate, about 3.7 in the total length; more distinct lateral band in fresh material. Head in total length 6.2; width of head in length 6.4—6.7; width of mouth in