Japanese Journal of Ichthyology V 魚 類 学 雑 誌 ol.18,No.2 September 15,1971 18巻2号1971年9月15日

Survival of Intergroup Percid Hybrids

Clark Hubbs (Received February 22,1971)

Abstract 264 Intertribial percid hybrid tests were made among the Percini,Luciopercini and Etheostomatini.Only those between the Luciopercini and Etheostomatini(both reciprocals)fed, indicating a close phylogenetic relationship.A series of morphologic characters also support separating the Percini from the other two tribes of North American .

Fish eggs can be fertilized by sperm of related Stizostedion had severely deformed larvae species and most of the putative natural hybrids (Kammerer,1907;Balon,1956)in striking are between relatively closely related species contrast to the putative"normal"natural (Hubbs,1955).The paternal and maternal hybrid reported by Peippo(1962). chromatin of relatively distantly related taxa Collette(1963)discussed the various methods may interact in the environment of the egg of arranging the percid genera and the fol- cytoplasm to initiate development and the lowing summary is based on his tribal group- amount of successful ontogeny is reasonably ings.The darters(Etheostomatini)are a closely associated with the accepted phylo- nearctic compact group of small benthic genetic relationship(Hubbs,1967a).The fishes comprising approximately 100 species relationship of hybrid survival to morpholo- in three or more genera,,, gically based phylogeny has been shown for and .The walleyes(Luciopercini) cobitids(Suzuki,1957),cyprinids(Suzuki, are a group of 5 closely related large piscivores 1968),centrarchids(Hester,1970),cyprino- usually placed in one (Stizostedion). donts(Hubbs and Drewry,1959),salmonids The (Percini)are a group of 7 species (Simon and Noble,1968),and many others. in 3 genera,Perca,,and The overall concordance of hybrid experiments Percarina.The fourth group is an eurasian and classification(Nikoljukin,1952;Kry- group(Romanichthyini)of four species in zhanovsky,1968)permits application of hybrid two genera,Zingel and . survival to help in determining phylogenetic This paper reports on the survival of hybrids relationships when morphological data are between species representing the first three conflicting. tribes and of representatives of those percid Many percid hybrids have been produced tribes with a variety of available species.The artificially.Hybrids between Stizostedion vit- three groups have been arranged with the reum and S.canadense have been reared by following relationships:perches and walleyes Nelson(1968),hybrids between Perca flu- in one group contrasted with darters(Jordan, viatalis and Gymnocephalus(=Acerina)cernua 1923);perches and darters,in one group have been reported by Kammerer(1907), different from but related to walleyes(Collette, and a series of crosses among darters reported 1963);and the three groups of equal taxonomic by Hubbs(1967b)who also reported less rank(Hubbs and Lagler,1958).The hy- success in interfamilial crosses.Despite all bridization results support the last interpre- of the tests on percid hybrids,none have tation but are in best accord with a grouping been done between darters and other percids of darters and walleyes contrasted with the and laboratory tests between Perca and and its allies.

•\ 65•\ 魚 雑Jpn.J.Ichthyol.18(2) September 15,1971

Materials and Methods as were all the 1969 tests.The rest were

done in Austin where the incubation tem- The gametes were removed from the parents and the eggs and embryos incubated following peratures available were 9.5•Ž,12.5•Ž,and 16.5•Ž.Eggs were placed in enamel rearing techniques developed by Strawn and Hubbs pans at the start of each test.Each lot of (1956)as modified by Hubbs(1971).Yellow Perca and Stizostedion eggs exceeded 100 perch eggs adhere to each other in large and frequently a lot was split as development one-layered mats and strings.After fertiliza- tion the eggs were divided into smaller units progressed and rearing space became available. Often some of the eggs were moved from a by cutting with scissors.The ripe adults cold to a warm incubation temperature. were maintained in sexual isolation to reduce

spontaneous mating.Stizostedion females re- The stocks used came from the following tained their eggs until the eggs were stale sources:Perca flavescens,Saline hatchery,

but egg strands extruding from Perca females Michigan Department of Conservation;Sti- were used successfully in hybridization tests. zostedion vitreum,mouth of Thames River,

Stizostedion males could be used successfully Tilbury,Ontario,Canada;Percina sciera, on the day following depletion of sperm Old Town on San Marcos River,Hays County,

supply by stripping but Perca males could Texas;Etheostoma spectabile,Old Town on not be reused once the sperm had been de- San Marcos River,and Big Maries River at Westphalia,Missouri;Etheostoma caerulum, pleted. The first series of tests were done April 1, E.blennioides,E.tetrazonum,and E.zonale, Big Maries River;Gambusia affinis,Clear 1969.The crosses were made at the portable laboratory of the Texas Parks and Wildlife Creek,16 kilometers west of Menard,Menard

Commission on the shores of Lake Meridith, County,Texas;Fundulus chrysotus,Martin near Fritch,Hutchinson County,Texas. Dies State Park,Jasper County,Texas;

Hybridization was done at lake water ambient Platypoecilius helleri,aquarium stock;Lepomis temperatures(ca 10•Ž)and the eggs returned macrochirus,San Marcos River in San Marcos,

to Austin in petri plates in styrofoam coolers. Hays County,Texas;L.microlophus,Martin The temperatures were 15•Ž or less upon Dies State Park;Chaenobryttus coronarius,

arrival in Austin where the eggs were placed farm pond 6 kilometers northeast of Manor, in enamel rearing trays at 15•Ž.Because Texas;Lepomis punctatus,San Marcos River

of instrumentation problems,the room tem- in San Marcos,Hays County,Texas;L. cyanellus,San Marcos River in San Marcos; perature soon reached 20•Ž and remained there for the rest of the test interval. Haplochromis,aquarium stock;Cichlasoma

The stocks used came from the following cyanoguttatum,San Marcos River in San Texas localities:Stizostedion vitreum,Lake Marcos.

Meridith at Fritch,Hutchinson County;Per- Two different types of indices may be used cina caprodes,Etheostoma lepidum,E.spec- to indicate hybrid success(1)duration of

tabile,Notropis venustus,Dionda episcopa,and hybrid survival and(2)development stage Lepomis auritus,Llano River at Junction, achieved by the hybrids.Usually the two

Kimble County;Esox lucius,stock tank near factors are similar but when discordant the Lake Meridith,and Morone chrysops,Lake former has recently been discounted.Slight

Nasworthy 5 km southwest of San Angelo, differences in experimental conditions(typically Tom Green County. temperature)can cause vastly different develop-

The second series of tests were done in mental rates in percids(Hubbs,et al.,1969)

April,1970.A few darter eggs were stripped and survival times if the hybrids and/or in Ann Arbor and held 1-2 days at 10•Ž controls were to die at a specific ontogenetic

and returned to Austin in styrofoam containers stage but be reared at different temperatures.

•\ 66•\ Hubbs:Survival of Percid Hybrids

Even without this distortion a time emphasis embryonic development.All 3 tests died is hazardous.As early as 1910 Moenkhaus on day 9 or 10.Walleye eggs were exposed reported that hybrids often had reduced to Esox lucius sperm in 10 tests.Abnormal developmental rates and Hubbs and Drewry embryos were noted in 7 tests on the fourth (1959)showed that the delays were often day,and in 4 of these tests none of the 2,3, associated with reduced survival.Finally, 5 and 8 embryos cephalized and all were Blair(personalcommunication,1970)and dead by the 8th day.The other 3 tests had Pierce(personal communication,1970)have one(from 4 embryos),2(from 12 embryos), reported bufonid and hylid hybrids that re- and 4(from 45 embryos)cephalized embryos mained in larval stages for more than 1 and 2 on the fifth or sixth days.The 4 in the last years respectively after the controls meta- test developed pigmented eyes on a head morphosed.Retarded developmental rates sunk into the yolk mass.No heart was would be expected of which had con- observed and all embryos were dead by the trasting(and conflicting?)genetic messages;or ninth day. hybrids would be predicted to slow develop- Tests of Stizostedion vitreum eggs with ment when their parental genomes were percid sperm were more successful.All 3 distinctly different.Therefore,I have used control tests had embryos on the third day, the developmental stage achieved as an index had tail free embryos on the fourth day and of hybrid success even though percid hybrids pigmented eyes(2,15,and 12)on the fifth and controls had comparable developmental day.Hatching(87%)occurred on the seventh rates at the same temperature. through ninth days and the last young died on the 12th to 13th days,apparently from Results starvation.The 3 tests with Etheostoma Three groups of eggs were used in the 1969 spectabile sperm all gastrulated on the third tests,one series(Esox lucius)provided no day,had tail free embryos on the fourth day, useful data because all experiments failed to and had pigmented eyes(29,62,and 152) gastrulate.This may be due to the extraor- on the fifth.The first 2 listed tests had only dinarily high temperatures(15-20•Ž)used for one and 2(respectively)normal embryos incubation or because the female used may hatched but the last had 107 mostly abnormal have been egg bound.The other eggs were young.The test with 2 embryos was ended all percids and results were in accord with on the 13th day but the other 2 lasted until current classifications. the 28th.The 3 tests with E.lepidum sperm Stizostedion vitreum eggs were exposed to all gastrulated on the third day,had embryos visible quantities of Dionda episcopa sperm on the fourth,and had pigmented eyes(18,

(3 tests with 3 males)and both mashed testes 16,and 22)on the fourth or fifth days.Only from a Lepomis auritus and in none of the 3 embryos hatched,2 abnormal from the 5 tests was any gastrulation noted.Walleye second and one normal in the last.The eggs were also exposed to Morone chrysops abnormal young were dead on the 11th day sperm in 3 separate tests with 3 males.All but the other lived 28 days.The 3 tests with 3 had gastrulated eggs on day 3,abnormal with Percina caprodes sperm gastrulated the embryos on day 4,and pigmented eye(2,5, third day,had embryos the fourth day and and 11)on day 5.Only one young with a pigmented eyes(22,22,and 94)the sixth day. stubby tail hatched.Its heart was beating The first had none hatch,the second 17 but it did not swim off of the bottom.Two mostly abnormal hatch the last of which died others broke the egg shell but died with their on the 25th day and the last had 10 normal heads still in the shells.The others had young that lived until the 28th day.All sunken heads and died without advanced darter-walleye combinations had some feed-

―67― 魚 雑Jpn.J.Ichthyol.18(2) September 15,1971 ing larvae. one egg hatched but it lived until its yolk The reciprocal tests were parallel.Controls sac was absorbed as did those maintained of E.spectabile were similar to those reported at 9.5•Ž.The developmental rate at 16.5•Ž by Hubbs(1967b),but are listed here because was notably faster than that at 9.5•Ž,—i.e. of the unusual thermal stresses involved. the rate for embryo formation was 2/3rds They gastrulated the third day,had embryos and for later stages about 1/2(Table 1). the fourth,had pigmented eyes the sixth and Obviously the effects of temperature on de- 90 %+ hatched on the eighth and ninth days. velopmental rate are greater just before hatch- Five of the 6 tests of E.spectabile eggs exposed ing than just after fertilization.Because only to S.vitreum sperm developed(the other had one test survived to the end of the experimental non-spherical=immature?eggs).They gas- interval,the duration of survival may provide trulated the third day,had embryos some with the best index for determining optimal con- their tails free on the fourth day and pigmented ditions.Use of absolute time would ignore eyes(14,21,16,15,and 29)the fifth day. thermal effects so a rate calculation was used 95% hatched and most seemed normal.The that had one day at 16.5•Ž equivalent to two first had all dead on the tenth day and the days at 9.5•Ž and 1.5 days at 12.5•Ž.Most last survived until the 21st day while the other transfers do no better than those held at 3 were gone on the 17th but had fed.The 3 constant temperature.The few that did tests with E.lepidum eggs gastrulated the were too scattered to provide a distinct pattern third day,had embryos,some with tails free, but 12.5•Ž seems to be a better temperature the fourth day,and had pigmented eyes(83, than either 9.5 or 16.5•Ž. 17,and 50)on the fifth or sixth days.All The survival of the Stizostedion controls hatched and the last young died on the 13th, shows that there is a racial difference in 16th,and 17th days respectively,the last two maximum developmental temperature tole- tests included young that fed.Only 3 of rance.Eggs from Lake Meridith,Texas the 6 tests with P.caprodes eggs developed (introduced from Iowa)develop at 20•Ž (the other 3 used eggs from the same female). whereas eggs from the Thames River,Ontario, The 3 successful tests gastrulated the third have trouble at 16.5•Ž. day,had tails free the fourth day,and had The walleye eggs incubated but not exposed pigmented eyes(22,33,and 14)the fifth day. to sperm did not gastrulate. 93% Hatched and the last embryos died the Walleye females were crossed with yellow

13th,14th,and 15th days respectively. perch males each with copious supplies of The Stizostedion vitreum control young milt 9 times,4 at 9.5•Ž and 5 at 16.5•Ž. swam toward the surface upon hatching Two of the 9.5•Ž samples had subsamples and were often free swimming the second day. transfered to 16.5•Ž.None of the 1000+ The Etheostoma spectabile control young eggs gastrulated. remained on the bottom for one week after Walleye eggs were exposed to darter sperm hatching.Both reciprocals of the hybrids 8 times.The 2 incubated at 16.5•Ž did not might be free swimming but each hybrid gastrulate but all of those at 9.5 did.The would settle to the bottom for prolonged in- 2 tests with E.blennioides sperm gastrulated tervals—the S.vitreum controls only did this the sixth day,had tails free on the 10th and when dying. 12th days,had eyes pigmented the 14th and

In 1970,the Stizostedion vitreum controls 15th days and hatched the 21st through 23rd were started at 9.5•Ž(4 times)and 16.5•Ž days(5 of 9 with pigmented eyes).The young

(4 times).Those exposed to the cold incuba- lived until the 35th to 37th days.Both tion temperatures were much more successful series had been split and transfered to 12.5•Ž than those placed in hot temperatures.Only on the 9th day.Five of 6 eggs with pigmented

―68― Hubbs:Survival of Percid Hybrids

Table 1.Survival of young Stizostedion maintained at hatching temperatures.

eyes hatched and one survived the experimental All of the young walleye darter hydrids were period.Three were observed to have large recorded as being on the bottom about half quantities of brine shrimp in their guts and of the time and free swimming the other half to have grown in length and bulk.Walleye of the time. eggs were exposed to E.caeruleum sperm Walleye eggs were also exposed to sperm 3 times.At 9.5•Ž the eggs gastrulated the of the following species:Fundulus chrysotus

sixth or seventh days and had tails free the (3 times),Gambusia affinis(2),Platypoecilius tenth day,had eye pigmentation the 13th helleri,Cichlasoma cyanoguttaturn(2),Hap-

day and hatched the 17th day.The survival lochromis,and Lepomis microlophus(2).None was low(only one of 5 eggs with pigmented of the eggs gastrulated. eyes hatched)but the hatched young lived The egg complements of 7 darters were

30 days.One lot of walleye eggs was exposed exposed to walleye sperm.All were incu- to E.spectabile sperm.The eggs gastrulated bated at 16.5•Ž.Only 3 had gastrulation.

the sixth day,had tails free the ninth day, The test with E.blennioides eggs gastrulated

pigmented eyes the 12th day,and hatched the third day,had pigmented eyes the seventh the 22nd day.The hybrid lived until the day but did not hatch.The 2 tests with P. 32nd day.This lot was also split and the sciera eggs gastrulated the third day,had tails

sublot incubated at 12.5•Ž.Two eggs hat- free the fifth day,had eyes pigmented the ched the 16th day and both fed on brine shrimp sixth day and hatched the twelfth day.The

but died the 47th day. only young that hatched lived 20 days.

―69― 魚 雑Jpn.J.Ichthyol.18(2) September 15,1971

Controls of Perca flavescens were done at interpreted to have uniparental origin. 9.5,12.5,16.5•Ž and all possible increasing Yellow perch eggs exposed to copious temperature combinations.It soon became supplies of walleye milt were incubated at

obvious that developmental rate was inversely 9.5•Ž(13),12.5•Ž(9),and 16.5•Ž(3).None correlated with egg number in the incubation of the eggs in the 16.5•Ž tests gastrulated. chamber,and that development was virtually The rates for 9.5•Ž are slower than those nil at 16.5•Ž(only 2 of 6 tests gastrulated at 12.5•Ž(in parentheses):gastrulation oc- and no eggs lived longer than 5 days incuba- curred after 4 or 5(3 or 4) days;eyes formed tion).The developmental rate at 12.5•Ž in 8 to 10(5 to 6)days and pigmented in was more rapid than that at 9.5•Ž as shown 17(12)days,tails were free in 7 to 11(6 to 7) in parentheses,gastrulation occured after 4 days and straight in 17 days.None of the

(5 to 6)days,eyes formed in 5 to 6(6 to 8) young at 12.5•Ž developed straight tails and days and pigmented in 11 to 12 (16)days,and none of those in either temperature were the tail was free in 7(10)days,straight in free swimming as also occurred in the 20

11 or 12(16)days.Three of the 7 samples other tests in which incubation temperatures at 12.5•Ž fed as did one of 9 at 9.5•Ž. Several were changed.Likewise none of the young samples were transferred to 16.5•Ž and those were observed to feed and all but one young that had gastrulated usually developed pig- died with large yolk sacs. mented eyes and had straight tails.Two Yellow perch eggs were exposed to E. sets of eggs transferred from 9.5 to 12.5•Ž spectabile sperm numerous times.None of fed as did one each of those transferred from those incubated at 16.5•Ž gastrulated.Those 9.5 to 16.5•Ž and 12.5 to 16.5•Ž at about at 9.5•Ž developed slower than those at

gastrulation. 12.5•Ž(in parentheses):the eggs gastrulated Yellow perch eggs are imbedded in a large in 4 or 5(3 or 4)days, the eyes developed in jellatinous mass and may break out at about 7(5)days and pigmented in 15 or 16(10 or 11) the time the eyes form or wait until after days, the tail was free in 7(6)days and straight they become pigmented.Early(premature?) in 12(10)days. Most young hatched normal- young do not move and seldom survive to ly;however all young incubated at 9.5•Ž swim free in the water.Successful young (21 pans),12.5•Ž(18 pans)and at increasing free swim at about the time the eyes pigment. temperatures(63 pans)failed to feed.Yellow

Nine lots of Perca flavescens eggs were perch eggs were also exposed to P.sciera

placed in incubation chambers without being sperm 4 times.At 9.5•Ž the eggs gastrulated exposed to sperm. Four contained eggs in after 5 days(3 or 4 at 12.5•Ž);the eyes de- which gastrulation occurred. The first lot veloped after 7(5 at 12.5•Ž)days and pig-

was suspected of being contaminated due to mented after 14(10 at 12.5•Ž)days,the tail sloppy technique but the others were done was free after 8(7 at 12.5•Ž)days and straight with care before any males were stripped after 13(10 at 12.5•Ž)days.Most young that day.Three of the tests were at 9.5•Ž hatched normally but none were observed

and the other at 12.5•Ž.The developmental to have fed in the 18 incubation pans.Yellow rates for 9.5•Ž(12.5•Ž in parentheses)were perch eggs were exposed to E.blennioides gastrulation in 8(7)days,eyes in 10 days, sperm 5 times.The one incubated at 16.5•Ž but never pigmented,and tail free in 11 to did not gastrulate but those at 9.5•Ž gastrulat- 13 days. None of the young were observed ed in 4-5 days,developed eyes before the to have a heart beat and all had the tail curled tenth day and had eye pigmentation by the down over the yolk mass. All of the embryos 16th day,and the tail was free in 11 days and developed at rates significantly slower than straight in 16 days. Many hatched in the those of the"fertilized" controls and are 4 held at constant temperature and the 4

―70― Hubbs:Survival of Percid Hybrids subsets transferred to 12.5•Ž,but none fed. pigmented eyes but none developed further. Yellow perch eggs were exposed to E.caeruleum The single test with E.zonale eggs did not sperm 4 times.Only one of the three tests gastrulate.Seven of the 8 tests with E.spec- at 16.5•Ž gastrulated and that single egg tabile eggs had some gastrulate and of these developed no farther.The one at 9.5•Ž had 68% developed pigmented eyes in 6 to 7 days. eye pigmentation and heart beat but did not One egg hatched in 8 days but the young was have a straight tail and did not feed;the sub- dead,10 others had a heart beat. set transferred to 12.5•Ž also did not feed. Discussion Yellow perch eggs were exposed to E.tetra- zonum sperm once at 9.5•Ž and the sample Hybridization tests may be plagued with subdivided with one transferred to 12.5•Ž. problems associated with exclusive maternal Both lots developed pigmented eyes and heart development.Usually I have been concerned beat but had curled tails and did not feed. with stimulation of the egg by heterospecific Yellow perch eggs were exposed to sperm sperm that did not contribute to subsequent from Lepomis macrochirus,L.cyanellus,L. ontogeny.The yellow perch tests showed punctatus,and Chanenobryttus coronarius. another method by which this problem could Embryos formed in all but the L.macrochirus develop.Several sets of yellow perch eggs tests.Those tests incubated at 16.5•Ž failed not exposed to any sperm gastrulated and to gastrulate,but those at 9.5 and 12.5•Ž many proceeded to late embryonic stages. developed embryos in 5 or 6 days(9.5•Ž) The first example was suspected to involve and 4 days(12.5•Ž),the tails were free in contamination through sloppy techniques,but 10(9.5•Ž)and 6(12.5•Ž)days,and straight later tests were done to reduce this potential in 17(9.5•Ž)and 8(12.5•Ž)days,the eyes to virtual zero.The females had been main- formed in 8(9.5•Ž)and 5(12.5•Ž)days but tained in the laboratory,isolated from males none pigmented in any of the 3 possible for 10 to 18 days.The first eggs removed combinations. from the female were used for the nonfertili- Yellow perch eggs were exposed to Ci- zation experiment so that contamination could chlasoma cyanoguttatum sperm 3 times and not come from that series of tests(Lindroth, only one lot gastrulated,but it did not pro- 1947,showed that Perca fluviatilis sperm gress farther. viability is measured in minutes),and the Darter eggs were exposed to copious sup- previous tests preceeded these by as much plies of yellow perch semen several times.All as 4 days.Contamination during experimental were incubated at 16.5•Ž,because most darter procedures is,therefore,essentially impossible. eggs do not survive at lower temperatures If the eggs develop from uniparental sources,

(Hubbs,1967b).The test with Percina sciera two alternative explanations remain.The eggs had 8 gastrulated eggs on the fourth day female may have had an ovotestis(reported and 3 embryos with pigmented eyes on the sixth in Perca by Brunelli and Rizzo,1928,and day but no farther development.Six of the widely occuring in the related serranids as 7 tests with Etheostoma caeruleum had some discussed by Smith,1965),or the eggs may have eggs gastrulate but only one embryo developed developed without any sperm.The latter alter- pigmented eyes and it died the next day.All native is supported by the developmental data. 4 tests with E.blennioides eggs had some eggs All stages took longer to develop than they gastrulate but only one test had any(10) did in the fertilized controls and survival embryos develop pigmented eyes.None was exceedingly low.Although these circum- hatched or developed a heart beat.All 3 stances could occur in self-fertilized eggs, tests with E.tetrazonum eggs had more than the degree of expression is more likely to be 30 eggs gastrulate,of these 41% developed extreme in functionally haploid development.

―71― 魚雑 Jpn.J.Ichthyol.18(2) September 15,1971

The exact causes were not an objective of results is likely to extend to the reciprocal this study,but the phenomenon relates to experiment as Balon and Kammerer reported interpretations of hybrid experiments.Tem- only on tests with Perca eggs and may have perature shock seems not to be involved be- attributed the absence of development of cause most of the tests were run at 9.5•Ž, Stizostedion=Lucioperca eggs to technique the temperature at which the parents had difficulties.*Our results are strikingly dis- been maintained for up to 18 days.All cordant with the putative hybrid reported hybrid tests were no more successful than by Peippo(1962).Because the specimen those of nonfertilized controls,may have been has not been located recently(Ryder,Koli, based on that type of development. and Toivonen,all personal communications, None of the eggs in the numerous similar 1971),the specimen has not been re-examined. nonfertilized tests with darters and walleyes The specimen was contrasted with kuhan gastrulated.More than 2000 darter eggs (=Stizostedion lucioperca)and ahvenen(=Per- (Hettler,1970)and 500 walleye eggs without ca fluviatilis)and seemed intermediate in exposure to sperm were isolated and incubated morphology.The soft dorsal rays were em- in a manner similar to the perch eggs. phasized(13-14 in P.fluviatilis;21-22 in The close affinity of darters and walleyes S.lucioperca,and 17 in the specimen).Ap- is supported by all of the hybridization data. parently S.volgensis was also examined as Although walleye eggs fertilized by darter an alternative explanation;however,both sperm usually had severe prehatching mor- S.canadense and S.marina may have 17 soft tality,the reciprocals often had more than dorsal rays and the putative hybrid could 90% hatching.The young(both reciprocals) merely be a S.marina that migrated to the were vigorous and fed on available food Baltic.**Because the hybrid nature of this supplies and could be reared through critical fish is discordant with all experimental evidence early feeding stages as well as or better than and other alternate explanations are available, the walleye controls. I consider that the putative hybrid is a mis- The experiments with yellow perch and identified Stizostedion. darters were less successful.Only one of Yellow perch x walleye hybrids have dras- the 21 darter complements(5 species)fertilized tically different results dependent upon the by yellow perch sperm had any hatching. reciprocal tested.Walleye eggs crossed with The single hatched young died less than 24 yellow perch sperm do not gastrulate;the hours after hatching.No swimming was reciprocal tests typically gastrulated and the observed.The reciprocal experiments were development was approximately parallel to more successful.Many young hatched and that of the controls until tail free stage,after appeared normal.They developed most at- which development slowed.Those young tributes at about the time the characters that hatched remained on the bottom and did appeared in the controls but typically stayed not swim unless the incubation pan was severely on the bottom after hatching whereas the agitated.No feeding occured similar to that maternal controls were free swimming.All common in walleye darter hybrids. hybrids(136 incubating lots)died prior to Walleye eggs exposed to serranid sperm feeding,but many lived until the yolk sac did develop.The development was distinctly was absorbed. less than that shown by walleye darter hybrids, The yellow perch x walleye hybrids had but far better than that of the walleye eggs many vertebral deformities such as the *Balon(personal communication ,1971)confirms severe dorsal bending of the vertebral column this hypothesis. figured by Balon(1956)and reported by **Peippo(personal communication ,1971)reports Kammerer(1907).The agreement of our that S.marina was not compared with the hybrid.

―72― Hubbs:Survival of Percid Hybrids

tested with Perca sperm.These results parallel equivalent taxonomic ranking,or preferably the the success of darter serranid hybrids(Hubbs, darters and walleyes to be grouped together.

1967a).Unfortunately,no serranid males The grouping of Perca and Stizostedion were available to test with the Perca eggs, seems to be based on Jordan's(1923)state- and no tests have been done using serranid ment about the Etheostomatidae"They differ eggs. from the Percidae in having six branchiostegals

Yellow perch eggs exposed to centrarchid instead of seven,the head unarmed and the sperm develop farther than the nonfertilized airbladder obsolete or nearly so."The air- controls and the fraction to gastrulate was bladder of darters varies in its development typically above 50%,whereas less than 10 and that variation(as yet not reported in of the nonfertilized controls produced em- detail in the literature)seems to be as great bryos.This hybrid survival shows that yellow as the variation between darters with large perch and sunfishes have a genetic similarity. air-bladdersand Stizostedionor Perca.Col- Darter eggs fertilized by sunfish sperm reach lette's report that Perca and Stizostedion anequivalent developmental stage(Hubbs usually have 7 but occasionally 6 and darters and Strawn,1957).The failure of the walleye usually have 6 but rarely 7 branchiostegals eggs exposed to Lepomis sperm may be an is in contrast to Jordan's use of that attribute artifact but is parallel to the Perca sperm to separate the taxa.The armored head(and results. large mouth)are attributes common in large

All tests with very differently related taxa piscivorous fishes but unexpected in small failed.The failures may be due to phylogeny arthropod eaters and could be merely con- or technique problems.For example the vergence on a similar adaptive peak.Because Cichlasoma•\ercid hybrids might be unable the Perca x Stizostedion hybrids have more to survive at low temperatures(Cichlasoma difficulties than either other intergroup com- lethality)and high temperatures(Stizostedion bination,the validity of this grouping is ex- or Perca lethality).Similar problems could ceedingly dubious.In contrast,the morpho- ensue with many of the other tests attempted. logical attributes cited linking them,seem

The genera of percid fishes have been equivocable. grouped into several subfamilial units.If The grouping of Perca and darters was we exclude the genera(Zingel and Romani- proposed by Collette(1963)who summarized chthys)not incorporated into hybrid studies, the Percinae"Percidae with the anteriormost the three groups,darters,walleyes,and perches interhaemal bone greatly enlarged;anal spines

have been arranged in three ways;(1)Perca usually large and well developed;lateral lines and Stizostedion(and the fishes now in Zingel) usually not extending onto caudal fin and

in one family Percidae and the darters in supplementary lateral lines on the caudal another the Etheostomatidae;(2)Perca and fin absent except sometimes in G.schraester." darters in the Percinae and Stizostedion The morphological basis of the grouping

(and Zingle)in the Luciopercinae;and(3) apparently is limited to three structures. Perca in the Percinae,Stizostedion in the Two are stated to be equivocable by Collette Luciopercinae,and darters in the Etheosto- which leaves one apparently unequivocable.

matinae.The first two arrangements listed Because the anterior interhaemals would be are discordant with the hybridization data. expected to support the anal spines,the size

Although the Perca x darter and Perca x of the two should be linked.If so,the develop- Stizostedion hybrids hatched,neither was seen ment of the interhaemals in darters with small

to ingest the food that the Stizostedion x anal spines might also be minimal and that darter hybrids ate and metabolized.Therefore, character also equivocable.Although Perca x the three groups should be considered to be of darter hybrids do better than Perca x walleye

―73― 魚雑 Jpn.J.Ichthyol.18(2) September 15,1971

hybrids,the present tests do not show develop- Acknowledgments ment to stages typically achieved by Sti- The 1969 experiments were done with the zostedion x darter hybrids;therefore,if two cooperation of the Texas Parks and Wildlife groups are to be allied,it is the last two. An extensive analysis of the morphological personel.I thank Marion Toole,Larry features of the three groups of percid fishes Campbell, and Jack Crabtree for their as- sistance in obtaining game fish stocks.B. found in North America is beyond the scope of this investigation,but some seldom used Jean Davis,Jana Velderman,and James Wiersema assisted in the hybridization ex- characters have been noted.Male Perca periments.The 1970 tests were done in flavescens has copious supplies of milt,but when that is expended no more could be conjunction with the Institute for Fisheries removed in the following two weeks;in con- Research,Michigan Department of Con- trast,males of Stizostedion vitreum and a servation. I thank W.Carl Latta for making large variety of darters could renew their semen arrangements to obtain Perca and Stizostedion supply even if held without food. Eggs of P. stocks.James Schneider for help in obtain- ing the stocks and arrangements for experi- flavescens are expelled in large numbers and adhere to each other and not to the substrate, mental space in Ann Arbor,and Mr.Glen in contrast,eggs of Stizostedion vitreum and Archibald of Tilbury, Ontario for donation of the walleyes he had obtained in the Thames darters are expelled in small numbers and adhere to the substrate.Eggs of Perca River.Missouri and Texas stocks were ob- tained through the courtesy of William S. flavescens are enclosed in a soft gelatinous mass,whereas Stizostedion vitreum and darter Birkhead,Elsie Boecker,Thomas Buchanan, and Gary Powell.I wish to thank Dr.R.A. eggs are enclosed in a hard chorionic shell. Developing embryos of Perca flavescens are Ryder for calling my attention to Peippo's typically capable of free swimming prior to paper.This investigation was supported by eye pigmentation,whereas embryonic Sti- NSF GB 6429. zostedion and darters have pigmented eyes long before normal hatching,and if hatched Literature Cited prematurely before eye pigmentation,are Balon, Eugeniusz. 1956. Medzidruhova hybridizacia incapable of locomotion. The last attribute dunajskych ryb I. Oplodnenie ikier ostrieza duna- of Perca is similar to conditions prevalent jskeho spermiou zubaca obycajneho (Perca fluviatilis in Lepomis ontogeny and Perca x Lepomis infraspecies vulgaris [Schaffer 1759] Pokrovskij hybrids develop to late ontogenetic stages. 1951 x Lucioperca lucioperca [Linne 1785]d'). Polnohospodarstvo, 5 : 581-592. Conclusions Brunelli, G., and L. Rizzo. 1928. Ghizndala esocrina, ovario impari ed erma froditismo nella" Perca Survival of intergroup percid hybrids sug- fiuviatilis". Atti. R. Accad. Naz. Lincei Cl. Sci.-is. gests that Stizostedion and darters are at Mat. et Nat., 7: 865-867. least as closely related to each other as either Collette, B. B. 1963. The subfamilies, tribes, and group is to Perca. Therefore,if two groups genera of the family Percidae (Teleostei). Copeia, 1963: 615-623. are to be allied,the walleyes should be placed Hester, F. E. 1970. Phylogenetic relationships of in the ,and Perca and its sunfishes as demonstrated by hybridization. Trans. allies would be the only tribe in the Percinae. Amer. Fish. Soc., 99: 100-104. The Etheostomatinae would then be divided Hettler, W. F., Jr. 1970. Effect of paternal death into three tribes, the Etheostomatini, Lucio- on sperm viability in the . Prog. Fish. Cult., 32: 209-211. percini,and Romanichthyini. Hubbs, Carl L. 1955. Hybridization between fish species in nature. Syst. Zool., 4: 1-20. , and K. F. Lagler. 1958. Fishes of the

―74― Hubbs:Survival of Percid Hybrids

Great Lakes region. Bull. Cranbrook Inst. Sci., Nikoljukin, N. I. 1952. Mezvidova gibridizacija ryb.

(26): xiii -1-213pp. (revised edition). Saratov: 312pp. Hubbs, Clark. 1967a. Analysis of phylogenetic Peippo, Lauri. 1962. Kalataloudelliselle tutkimu-

relationships using hybridization techniques. Symp. stoimistolle lahetettyja kalanaytteita. Eripaimos Newer Trends in . Bull. Nat. Inst. Sci. Soumen Kalatuslehti, 5: 183-184.

India, 34: 48-59. •\ Simon, R., and R. Noble. 1968. Hybridization in

. 1967b. Geographic variations in survival Oncorhynchus (Salmonidae). I. Viability and in- of hybrids between etheostomatine fishes. Bull. heritance in artifical crosses of chum and pink

Texas Mem. Mus., 13: 72pp.•\ salmon. Trans. Amer. Fish. Soc., 97: 105-118. . 1971. hybridization studies. Proc. Smith, C. L. 1965. The patterns of sexuality and the

Calif. Acad. Sci., (fourth series) 38: 289-298. •\ classification of serranid fishes. Amer. Mus. Novit.,

, and G. E. Drewry. 1959. Survival of F1 2207: 2Opp. hybrids between cyprinodont fishes, with a discus- Strawn, K. and C. Hubbs. 1956. Observations on

sion of the correlation between hybridization and stripping small fishes for experimental purposes.

phylogenetic relationship. Publ. Inst. Marine Sci. Copeia, 1956: 114-116. Univ. of Texas, 6: 81-91. •\ Suzuki, R. 1957. Hybridization experiments in

, A. E. Peden, and M. M. Stevenson. 1969. cobitid fishes. Jap. J. Genet., 32: 8-14. The developmental rate of the , . 1968. Hybridization experiments•\ in cyp-

Etheostoma lepidum. Amer. Midl. Nat., 81: 182- rinid fishes. XI. Survival rate of F1 hybrids

188. •\ with special reference to the closeness of taxonomical

, and K. Strawn. 1957. Survival of F1 position of combined fishes. Bull. Freshwater hybrids between fishes of the subfamily Etheosto- Fish. Res. Lab., 18: 113-155. minae. J. Exp. Zool., 134: 33-62. Jordan, D. S. 1923. A classification of fishes including (Department of Zoology, The University of Texas at

families and genera as far as known. Stanford Austin, Austin, Texas 78712, U. S. A.) Univ. Ser. Biol. Sci., 3(2): 77-243.

Kammerer, Paul. 1907. Bastardierung von Fluss- Percidaeに お け るtribes間 雑 種 の 生 残 能 力 に つ い て barsch (Perca fluviatilis L.) and Kaulbarsch (Acerina Clark Hubbs

cernua L.). Archiv Entwicklungsmechanik Organis- ペ ル カ 科(Percidae)の 三 っ のtribes,す な わ ち, men, 23: 511-551, pls. 22-23. percini,Lucioperciniお よ びEtheostomatiniの 問 で の Kryzhanovsky, S. G. 1968. Zakonomernosti razvitija 交 雑 合 計264が 行 な わ れ た が,こ の 中 でLucioperciniと gibridov ryb razlichnysks sistematicheskikh kategorij. Moscow: 219pp. Etheostomatini(逆 交 雑 両 方 向 に つ い て)の 組 み 合 わ せ

Lindroth, A. 1947. Time of activity of freshwater の も の の み が 生 育 し た.こ の こ と は,こ の 両tribesが fish spermatazoa in relation to temperature. Zool. 系 統 的 に 近 縁 で あ る こ と を 示 す 。 一 連 の 形 態 学 的 形 質 も Bidr. Upsala, 25: 165-168. Perciniが 北 部 ア メ リ カ に お け るPercidaeの 他 の 二 っ の Moenkhaus, J. 1910. Cross fertilization among fishes. Proc. Indiana Acad. Sci., 1910: 363-390. tribesと は っ き り 区 別 さ れ る こ と を 支 持 し て い る.

Nelson, W. R. 1968. Embryo and larval characteris- tics of sauger, walleye, and their reciprocal hybrids. (アメ リカ合 衆 国 オ ー ス テ ィン テキ サ ス大 学 動 物 学

Trans. Amer. Fish. Soc., 97: 167-174. 教 室)

―75―