198 JouRrveloF THE At'lnRrceN Moseurro Cotrnol Assocrerror'l Vol. 7,

DIAGNOSTIC CHARACTERIZATION OF FREEBORNI AND AN. HERMSI BY HYBRID CROSSES,FREQUENCIES OF POLYTENE X CHROMOSOMESAND TDNA RESTRICTION ENZYME FRAGMENTS

G. N. FRITZ,I S. K. NARANG,' D. L. KLINE,' J. A. SEAWRIGHT,' R. K. WASHINO.I C. H. PORTER' lr.ro F. H. COLLINS3

U.S. Department of Agriculture, Agricultural ResearchSeruice, Medicat and Veterinary EntomologyResearch Laboratory, P. O. Box 14565,Gainesuille, FL 32604

ABSTRACT. A qolytene chromosomeanalysis was prepared ftom Anophelesfreeborni collectedfrom 25 Iocations in north and central California, and parts of Washington and Oregon.The X chromosome banding pattern, thought previously to be specifii to An. hcrmsi, *as co--otr-in mosquitoescollected from_foothill regions in California, and in all samplesfrom Washington and Oregon. Ai some of these lo-cation-s,many mosquitoeshad heterokaryotypesfbr the inversion thlt distinguishls the X chromosome of An. freeborni from that of An. hermsi. Use of rDNA restriction site analysis, and the results from crossingof different strains bearing either type of X chromosome,showed thai An. hermsi doesnot have a unique.or diagnosticX chromosome. Anopheleshermsi wascollected in San Mateo County, CA, which is now the northernmostknown limit of this species.Crossing studies, or the examinationof 1DNA restriction enzymeprofiles, are presently the only meansof identifying An. herrnsi.

INTRODUCTION cording to the description of An. hermsi (Barr and Guptavanij 1988),the only characterthat A new speciesof anopheline ,Anoph- distinguishesboth speciesis a simple inversion eleshermsi, was describedrecently by Barr and on the polytene X chromosome.The banding Guptavanij (1988). At present, An. hermsi is patterns on the autosomesof both speciesare thought to be limited to areasof southern Cali- either identical (Baker 1965,6Fritz 19897)or too fornia south of the Tehachapi Mountains (Cope similar (Menchaca19865) to be useful as diag- et al. 1988).This species has beenimplicated in nostic characters.Baker and Kitzmiller (1963) recentautochthonous cases of vivax in first describedthe X chromosomeof An. hermsi southern California (Maldonado "southern et al. 1990). (then referredto as occidentalis")and Prior to 1988, mosquitoes now known as An. noted that it had unmistaken homology with hermsi were believed to be a coastal strain of that of An. Both types of X chromo- Anopheles Aitken (Aitken freeborni. freeborni 1939,1945), some appearedto differ only by a single inver- a geographic variant of Anopheles occidentalis sion (In(X)A) at subzones2B-5C (Baker 19656). (Lewallen Dyar and Knab 1957)or a new species As a part of a study on the population genetics (Baker and Kitzmiller 1963, Fujioka 1986,a of An. freeborni,we examinedthe polytenechro- Menchaca19865). mosomes of specimens collected throughout present, At there are no reliable anatomical north and central California, and parts of Ore- charactersthat distinguish eggs,Iarvae or adults gon, Washington and Utah. The purpose of this of.An. hermsi from those of An. Ac- freeborni. cytogenetic survey was to distinguish An. free- borni from An. hennsi, record chromosomalpol- ymorphisms and use these polymorphisms as 1 Department of Entomology, University of Califor- genetic markers to determine population struc- nia. Davis. CA 95616. ture and speciesdistributions. ' USDA, ARS, Medical and Veterinary Entomology Our initial survey showed that the X chro- Research Laboratory, P. O. Box 14565, Gainesville, mosome banding pattern described as specific FL 32604. present 3 to An. hermsi was in some samples Malaria Branch Division of Parasitic Diseases, collectedin northern and central California. and Centers for Disease Control. Atlanta. GA 30333. nFujioka, parts of Oregon and Washington. Furthermore, K. K. 1986. Hybridization and electro- phoretic studies of 3 members of the North American Anopheles maculipennis complex (Diptera: Culicidae). Ph.D. dissertation, University of California, Los An- o Baker, R. H. 1965. Cytogenetic evidence for the geles. evolutionary relationships among the Nearctic rnocu- u Menchaca, D. M. 1986. The cytogenetic study of lipennis speciesof anopheline mosquitoes.Ph.D. dis- an undescribed member of the North American sertation, University of lllinois, Urbana. A nophe Ies maculipennis (Diptera: Culicidae) complex. 7Fritz, G. N. 1989. Mass rearing and population Ph.D. dissertation, University of California, Los An- genetics of Anopheles freeborni. Ph.D. dissertation, geles. University of Florida, Gainesville. PoLYTENEX Csnonaosoun

Table 1 Collection site number and location for samples of Anoph.elescollected in California, Oregon and Washington.

Site no. Location

California 1 NevadaCo., Wolf Creek& Highway 49 2 El DoradoCo.. Camino. Carson Rd. 3 El DoradoCo., Pleasant Valley Rd. 4 SacramentoCo., Sloughouse SacramentoCo., Folsom t) Sutter Co.,Highway 99 & HowsleyRd. 7 Yolo Co.,Capay Valley, Guinda 8 Yolo Co.,Knights Landing q ColusaCo., Millers Landing 10 Colusa Co., Highway 20 near Williams 11 Sutter Co.,west of Yuba City on Butte HouseRd. 12 Butte Co.,Chico IJ TehamaCo., Tehama, Gyle Rd. t4 Glenn Co.,east of Willows on Highway 162 15 Lake Co.,Clear Lake 16 SonomaCo.. Sonoma. Huichica Cr. 1a SacramentoCo., Highway 99 & Twin Cities Rd. 18 San Mateo Co.,Jasper Ridge Preserve on Sand Hill Rd. 19 Kern Co.,Onyx, CanebrakeCreek 20 Inyo Co.,Big Pine 2l Inyo Co.,Bishop Oregon 22 JeffersonCo., Madras oa Umatilla Co.,Hermiston Washington 24 Benton Co.,Richland 25 Yakima Co.,Yakima

many mosquitoesat someof theselocations were MATERIALS AND METIIODS heterokaryotypes (one copy of the standard X sarnplesand chrornosonxepreparation: chromosome specific to An. freeborni and one copy of In(X)A). These unexpected results Larvae and adults were collected during July- (Table raised some important questions: October 1988 from 25 different sites 1. 1) Was An. herrnsi more widely distributed Fig. 1). In California, many areas throughout than thought previously? the SacramentoValley were sampled as well as locationsin the foothills of the Sierra Nevada, 2) Were An. freeborni and An. hermsi really distinct species? the coastalrange,2locations in the OwensVal- ley and one location the coast at Palo Alto. 3) Were individuals of An. freeborni and An. near h,ermsi mating with one another in sympatric The techniquesdescribed by French et al. (1962) wereused to preparethe polytenechromosomes. populations? "C 4) Was one species,both, or perhaps an uni- Bloodfedfemales were kept at 28 for approx- dentified third speciespolymorphic for In(X)A? imately 28-29 h. Ovarieswere dissectedin dilute To answer these questions, we have crossed Carnoy's and then transferred to 75Vo acetic various geographic strains having one or the acid. Chromosomes,banding patterns and in- other type of X chromosome.According to Fu- versions were identified using a photographic jioka (19864),crossing An. freeborni to An. map made of the best polytene chromosome hermsi will produce sterile hybrid male progeny. preparations.This map was used in conjunction In addition. we examined rDNA restriction en- with the description and standard map of ovar- zyme pattern variations, which were shown re- ian nurse-cell polytenes (Faran 19818).The cently to distinguishboth species(Collins et al. 1990).We report here the frequencyand distri- bution of both types of X chromosomesfound 8Faran, T. C. 1981.The adult ovarian nurse cell in our cybogeneticstudy, the results obtained chromosomesof Anopheles(Arcpheles) freeborni Ait- from crossing different strains and the results ken 1939(Diptera: Culicidae). M.Sc. thesis, University of the rDNA analyses. of Maryland, CollegePark. 200 JounNlr,oF THE Ar',tenrclN Moseurro CoN.rRor,Assocrlrrox VoL.7,No.2

WASHINGTON

OYAKIMA

CALIFORNIA O MADRAS .,9TEHAMA @(t 50MILES OREGON O @(l A " 'i: ^"O' 'u1; (9. , ?"ril'

LOSANGELES

Fig. 1. Frequency of the standard (white) and inversion (black) karyotype for the X chromosome at collection sites in California, Oregon and Washington.

Marysville strain (which Faran usedto prepare ner describedby Collins et al. (1990). These her map) was obtained from the Walter Reed sites included those in which mosquitoes were Army Institute of Researchand usedas a stand- polymorphic for the inversion on the X chro- ard for comparing banding patterns. mosome,as well as those fixed for either homo- rDNA probe: The rDNA of An. freeborni and karyotype.Prior to the rDNA analysis,the po- An. hermsi have a restriction enzymesite differ- lytene X chromosomekaryotype of each female ence in the external and internal spacerregions mosquito was determined. Samples of An. (Collins et al. 1990). These diagnostic differ- h.ermsifrom 3 sites in southern California were enceswere usedto identify speciesin this study. alsoprobed. In a blind test, the rDNA of at least 3 individuals Crosses.'Crosses were made among several from each of 12 collection sites in California, geographicstrains presumedto be An. freeborni. Washington and Oregonwas probed in the man- To examine widely separatedgeographic strains JUNE 1991 AN. HEBMSTPoLYTENE X CgRouosolrp that also showedsome habitat and chromosomal RESULTS differences,a strain from Richland, WA (WASH strain, site 24,Table 1), was crossedto a strain X chromosome:In California, all mosquitoes from Davis (DAVIS strain) in the Sacramento from the SacramentoValley (sites4-14 in Table Valley, CA. Mosquitoes in the former site were 2, Fig. 1) and the Owens Valley (sites 20, 2L) polymorphic for In(X)A (i.e., included hetero- had the standard X chromosomedescribed by karyotypes),an inversion hitherto unknown in Kitzmiller and Baker (1963)and Faran (19818) An. freeborni.The DAVIS strain possessedthe as that of An. freeborni. Samples from the standard homokaryotype only. coastal mountain range or the foothills of the Three strains from California, which were Sierra Nevada, however, were either fixed for fixedfor oneor the other form ofX chromosome, In(X)A (type of X chromosomefound in An. were also crossed.These included: 1) JASP hermsi) or were polymorphicfor the inversion. strain (fixed for In(X)A), collectedat the Jasper At Clear Lake (site 15), for example, 8% of the Ridge Preserve (site 18) just west of Palo AIto mosquitoessampled were heterokaryotypes.At and ca. 16 km from the coast:2) LAKE strain Onyx (site 19), the frequency of heterokary- (also fixed for In(X)A), collectedat the north otypes appeared to be greater, but the sample end of Clear Lake (site 15) in Lake County; and size was too small for an accurate estimate of 3) DAVIS strain (standard homokaryotype). karyotype frequency.Jasper Ridge (site 18) and When crosseswere made, the WASH strain was Camino (site 2) were fixed for the inversion in its 9th laboratory generation, whereas the karyotype. Another sample taken from Camino LAKE and JASP strains were in their first. The (site 3) was unusual in that one standard hom- DAVIS strain had already been maintained in okaryotype was found among 12 inversion hom- the laboratory for over 4 years. okaryotypes. Parentalcrosses among the JASP, LAKE and AII mosquitoescollected in Madras, OR (site DAVIS strains were done by forced copulation 22), wete inversion homokaryotypes (Table 2, (Baker et al. 1962).The parental crossesbe- Fig. 1).As oneproceeded north into Washington tween the WASH and DAVIS strains and all State, the frequency of the standard karyotype other crosseswere done by combining virgin increased.At Yakima (site 25) the standard X mosquitoes same and each sex. of the age of chromosomekaryotype reached its highest fre- These were then allowed to mate freely in cages. quency of 0.90. The observedfrequencies of Bloodfeeding, handling of eggs and all other the homokaryotlpes and heterokaryotypes in rearing techniques were standardized and are Hermiston (site 23),Richland (site24) and Yak- describedin detail by Fritz (1989?).Egg batches ima (site 25) did not differ significantly from that did not hatch were held for at least 6 days those expectedunder Hardy-Weinberg equilib- and checkedfor the presenceor absenceof em- rium (Table 2). bryonic development. The spermathecaeof fe- parental males that laid unhatched eggs were dissected Crosses.'In the series of crossesbe- to determine the presenceor absenceof sper- tween the DAVIS and WASH strains, no signif- matozoa. Uninseminated females that did not icant differences were observed in fecundity, lay any eggs(n : 8) and those that died without fertility and the percentageof adult emergence Iaying eggs (n : 4) were excluded from any (Table 3). The sexratio (males/females)ranged statistical analyses. At least 10 hybrid males from 1.07 to 1.48, but none was significantly from each crosswere checkedfor the presence different (P>0.05). Although the 2 reciprocal of spermatozoain the testesand for the devel- crossesdiffered significantly from each other in opment of normal genitalia. the mean number of pupae produced, neither All parental crosses and backcrosses were cross differed from the controls. comparedwith one another and with controls Backcrossesdid not differ significantly from for fertility and fecundity. The polytene chro- controls in fecundity, fertility and percentageof mosomesof hybrid females were also checked adult emergence.The sex ratios ranged from for banding pattern homology and the degreeof 0.91to 1.36,but none differed significantly from synapsisbetween homologues. eachother (P>0.05).Ten of 12 backcrossespro- Basic summary statistics of the data were duceda significantly higher percentageofpupae done using SAS (Statistical Analysis Software). than controls; in effect, post-hatch mortality A one-way analysis of variance (ANOVA) com- was significantly lower among backcrosses.The pared mean values of the number of eggslaid/ controls in the backcrossseries differed signifi- female,the percentageofeggs hatched, the num- cantly from each other in percentagepupation, ber of pupaeproduced per eggbatch, the number although this difference was not evident in the of adults emergingand the sex ratio. controls done during the parental crosses. 202 JouRNar,oF THEAMERTcAN Moseurro CoNrRor,Assocra,rrou VoL.7,No.2

Table 2. Observed(o) and expected(e) numbers of standard homokaryotypes(S/S) inversion homokaryotypes (I/I) and heterokaryotypes(S/I) for In(X)A of Anophelescollected from various sites in California, Oieg-on and Washinston. ChromosomeX Freq. S/S S/I r/r chi Site" sq. 1 I 1 0 0 1.00 0.00 2ab 27 0 0 0 0 27 0 0 0.00 1.00 3ac AJ 1 0.08 0 1.85 72 11.08 t2.5I 0.08 0.92 4 I 9 q 0 0 0 0 0 1.00 0.00 5 0 0 1.00 0.00 b 50 50 50 0 0 0 0 0 1.00 0.00 7 3 3 0 0 1.00 0.00 8 22 22 0 0 0 0 0 0 1.00 0.00 q 35 35 35 0 0 0 0 0 1.00 0.00 10 50 50 50 0 0 0 0 0 1.00 0.00 11 22 22 22 0 0 0 0 0 1.00 0.00 t2 27 27 27 0 0 0 0 0 1.00 0.00 IJ 44 44 44 0 0 0 0 0 1.00 0.00 T4 50 50 50 0 0 0 0 0 1.00 0.00 15b 60 0 0.01 4.56 55 55.20 0.04 0.04 0.96 16 I I 0.00 1.00 t7 16 1t) 16 0 0 0 0 0 1.00 0.00 18a 41 0 0 0 0 4I 4T 0 0.00 1.00 19 /b 0 3 1 0.38 0.62 20 q I q 0 0 0 0 0 1.00 0.00 2l 0 0 1.00 0.00 22ab 15 0 0 0 0 15 15 0 0.00 1.00 23c 47 ft J.O I t4 18.83 27 24.57 3.06 0.28 0.72 24d. na 26 27.73 38 34.46 q 10.73 0.75 0.62 0.38 25e 50 4T 40.50 8 9.00 1 0.50 0.62 0.90 0.10 'Sample sites followed by the same letter do not differ significantly in frequenciesof S/S, S/I and I/I (homogeneitychi squareP : 0.05). 'Includes 2 femalescollected by Stan Cope.

The testesof the hybrid Fr maleswere similar the hybrid male progeny had no sperm in their in appearanceand amount of sperm as those of testes. Although the genitalia appeared to be controls,and all genitaliaappeared normal. The normal, the testes were often translucent and chromosomes of F1 hybrids synapsed equally smaller than those of the controls. Fr hybrid well as thoseof controls. males, from crossesin which the female parent In the parental crosses among the LAKE, was from LAKE or DAVIS, had varying DAVIS and JASP strains, the controls did not amounts of sperm in their testes. The amount differ significantly from the reciprocal crosses varied from none to quantities that appeared in fecundity, fertility, the percentageof individ- normal; the testes were filled with what ap- uals pupating or the percentage of individuals peared to be globular spermatocytes and par- emergingas adults (Table 4). The ratios of males tially developedspermatozoa. to femalesranged from 1.07to 1.48but were not The results of the backcrossseries (Table 4) significantly different among parental crosses confirmed the results obtained from the dissec- (P>0.05). tion of hybrid males. All crossesinvolving hy- All Fr hybrid males from crossesbetween the brid males,in which the parental male was from LAKE and DAVIS strains had genitalia and the LAKE or DAVIS, producedeggs that did quantities of sperm that were similar to those not hatch or contain any stage of embryonic observed in the controls. However, F, hybrid development. Most crosses involving hybrid males from crosses between the LAKE or males, in which the parental male was from the DAVIS strain and the JASP strain were com- JASP strain, had mean percentageegg hatches pletely or partially sterile. When the male parent that were significantly lower than those of the of the hybrid was from the LAKE or DAVIS controls; unhatched eggs in these crossesalso strain and the female was from the JASP strain. contained no embryos. JUNE 1991 Att. nsntwst Polvtpxr X CnRol,tosovr

Table 3. Mean numberu of eggs laid/female, percentb hatch, percentb pupation and percentb emergence of adults of crosses made between the DAVIS strain (D) and the WASH strain (W) of Anopheles freeborni. Eees/ % % % Cross" female hatch puparlon emergence

Experiment 1: Parental crossesand controls WxW l7 131a 70.2a 61.3ab 89.3a DXD t7 135a 78.9a 64.1ab 92.7a WXD 19 I40a 72.4a 51.9b 92.2a DxW 18 126a 66.3a 74.\a 92.5a Experiment2: Backcrossesand controls DXD 10 111ab 95.4a 62.4ef 93.9ab WxW I I25ab 96.4a 31.5g 94.9ab DWXD 10 97b 93.1a 83.1abc 90.3ab DWxW 10 153a 77.5a 83.2abc 91.3ab DWXDW 10 125ab 81.4a 81.7abc 89.8ab DXDW o 12Oab 92.7a 79.5abcd 87.4b WXDW 10 I29ab 90.9a 92.8a 92.2ab WDxW q 138ab 92.9a 81.9abc 93.9ab WDxD q 148ab 95.0a 83.5abc 94.1ab WDxWD 10 153a 94.9a 87.Oab 91.3ab WXWD 10 128ab 84.4a 66.8cdef 92.4ab DXWD o 1,52a 93.7a 75.6abcde 90.7ab DWXWD 10 93b 93.4a 90.8ab 96.2ab WDXDW l0 141ab 76.|a 86.9ab 97.4a n Means in the same column and experiment, followed by the same letter, are not significantly different (f>0.05). Separate ANOVAS were done for each experiment. r'Percentages refer to the proportion of individuaG moving from one life stage to the next. " The first letter of each cross represents the female Darent.

There were no significant differencesbetween DISCUSSION controlsand backcrossesin the meannumber of eggs laid per female. All hybrid females were As yet there are no sibling speciesof anophe- fertile and had similar mean percentagehatch Iines that, when crossed,do not produce some as controls when backcrossedto either parental degreeof sterility in the hybrid progeny. This is strain. Many of the backcrosseshad a mean true even when the speciesare homosequential percentagehatch that was higher than the con- in their polytene chromosomebanding patterns. trols, indicating possibleheterosis, but none of For example,An. atroparuusVan Thiel and An. these differenceswere significant (P>0.05). labranchiae Falleroni are homosequential spe- The polytene ovarian chromosomesof hybrid cies in the Palearctic Maculipennis Group and progeny did not differ from controls in the producesterile male hybrids (Bianchi 1968,Co- amount or degree of synapsis, or in banding luzzi and Coluzzi1969). pattern. The X chromosomesof hybrids between The degreeand causesof sterility among spe- the LAKE and JASP strains (both fixed for cies vary greatly, but it is generally true that In(X)A) synapsedcompletely. All hybrids from post-zygotic barriers exist between most sibling crossesbetween the JASP or LAKE strain with speciesof mosquitoes(see reviews the DAVIS strain (fixed for the standard kary- by Kitzmiller otype) were heterokaryotypes. 1953, Kitzmiller et al. 1967, Kitzmiller 1976. Narang and rDNA probe: Samplesof mosquitoesfrom 11 Seawright 1991). In such crosses of 12 collectionsites in California,Washington eggsmay not hatch, or larvae may only reach a and Oregon had the same restriction enzyme certain stageof development;sex ratios can also fragment pattern regardlessof their X chromo- be skewed, and adults can be malformed or somekaryotype (Table 5). Mosquitoesfrom Jas- sterile. In crosseswhere adults are produced, it per Ridge (site 18), and those from samples of is almost alwaysthe malesthat are sterile. Males An. h.ermsi collected in southern California. may be sterile in only one of the 2 reciprocal were the only individuals with a restriction frag- parental crossesor in both. Of the 30 possible ment pattern specificto An. hcrmsi. No individ- crossesbetween sibling speciesin the An. garn- uals had a restriction pattern that was a hybrid bioe complex, all but 2 produce sterile males of that found in An. herrnsi and.An. freeborni. (Davidson1964, Davidson and Hunt 1973). 204 JouRru, oF THEAunnlcnN Moseurro CoNrnorAssocrlrroN Vor,.7, No. 2

Table 4. Mean numberuof eggslaid/female, percentb hatch, percentb pupation and percentbemergence of "d"ttr for "r"..". -

Eggs/ % % % Cross" n female hatch pupation emergence

Experiment 1: Parental crossesand controls DXD b 146ab 81.3a 73.9a 91.3a JXJ a 119ab 64.2a 53.0a 87.4a LXL 118ab 79.6a 59.0a 88.6a DxL a 93b 86.1a 73.9a 100.0a DXJ 159a 93.2a 80.4a 98.3a LXD t lDban 76.4a 76.8a 87.9a JXD q 137ab 93.9a 71.2a 85.9a JXL I 126ab 82.2a 67.4a 89.9a LxJ 163a 87.5a 56.3a 81.8a Experiment 2: Backcrosses and controls Eggs/ Eggs/ Cross' female "/ohatch Cross" n female % hatch DXD 5 94ab 78.59abcd JDxD A 134a 99.1a LXL / 104ab 70.3abcd JDXJ o 94ab 62.2cd. JXJ o 100ab 55.1de LXLD 4 102ab 96.8ab DxDJ 8 l20ab 16.6f LXDL l42a 90.1abc DJXDJ b l42a 25.ief LDXL 8 l20ab 97.9ab JXDJ o 116ab 19.3f LDXD 4 l18ab 98.5ab JXLJ 3 64b 30.6ef LDXLD 6 l17ab 83.8abcd L xLJ a 119ab 3.3f DxDL o l22ab 96.9ab LJXLJ 6 128ab 10.0f DXLD D 98ab 82.1abcd DXJD 8 104ab 0.0f DLXL 67b 74.7abcd JXJD o 118ab 0.0f DLXDL l13ab 96.2abc JDXJD 4 l22ab 0.0f DLXD 4 141a 76.8abcd LXJL , 99ab 0.0f JLXJ 4 107ab 50.5ef JXJL I 95ab 0.0f JLXL l05ab 72.7abcd JLXJL K 90ab 0.0f LJXL 6 78ab 74.6abcd DJXD o 97ab 95.2abc LJXJ o 98ab 89.8abc DJXJ n l2Iab 95.1abc u Means in the same column, followed by the same letter, are not significantly different (P > 0.05). Separate ANOVAS were done for each experiment. b Percentages refer to the proportion of individuals moving from one life stage to the next. " The first letter of each cross reDresents the female oarent.

In our study, sterile hybrid progeny were pro- from a strain lacking a particular transposon ducedonly when the JASP strain was crossed (e.g.,P element in Drosophila)is crossedto a to either the LAKE or DAVIS strain. The cause male from a strain having the transposon. of sterility was due to a complete lack or small Therefore, dysgenesisoccurs generally in just quantity of developed spermatozoa in hybrid one of the 2 reciprocal cross-hybrids. Hybrid males.This pattern is identical to that found by dysgenesisis characterizedby substantially ele- Fujioka (1986")when he crossedAn. hermsi to vated rates of mutation, chromosomal re- An. freeborni. arrangementand illicit recombinationin males Cytoplasmic incompatibility between strains (Drosophila).Dysgenic sterility in Drosophilais due to symbionts (Barr 1980)or the movement usually more pronounced in females (Engels oftransposable elementsdoes not provide a good 1980). explanation for the sterility observedin crosses In our study, as in Fujioka's (1986n),sterility of the JASP strain with the LAKE or DAVIS was limited to males and was present in hybrids strain. Cytoplasmic incompatibility, as observed from both reciprocal crosses.Fecundity and fer- in natural populations of Culexpipiens Linn., is tility of hybrid females was not significantly maternally inherited and causessterility in the different from that of controls, suggestingthat parental generation; in effect, the parental fe- disruption of germline genetic and develop- male deposits an egg raft that generally fails to mental integrity (characteristic of hybrid dys- hatch. genesis)had not occurred.Consequently, genetic Transposableelements can causehybrid dys- differences between the JASP strain and the genesis, which usually occurs when a female LAKE or DAVIS strain, rather than transpos- JUNE1991 AN. nnnptst PoLvTENEX Csnorrrosouo 205

Table 5. Sample site, type of polytene X ifornia, Hardy-Weinberg equilibrium with re- chromosome,and the IDNA probedetermination of -- spect to X chromosomekaryotype does not ap- species.F freeborni type; H : hermsi type; HF : pear to be the rule. The collectionof one stand- - : heterokaryotyPe; not scored. ard homokaryotype among t2 inversion X IDNA X TDNA homokaryotypesin site 3 (Table 2), and similar Site chrom. probe Site chrom. probe resultsfrom recentcollections (G. N. Fritz. un- 2H F20 FF published data) suggestsome degree of repro- 2H F20 FF ductive isolation between both X chromosome 2H F22 HF karyotypes. Since there is no hybrid sterility or 6F F22 HF rDNA restriction pattern difference between 6F F23 HFF An. freeborni collectedin the Central and Owens 6F F23 FF valleys compared with those mosquitoes col- 12F F23 HF lected in foothill regions,we are, at present, 12F F24 HFF T2F F24 FF consideringthem as simply ecotlpesof one spe- 13F F24 HF cies. 13F F25 HFF The results from this study support the spe- 13F F25 FF cific designation of An. hermsi. Furthermore, 15 HF 25 HF An. hermsi was not found to be sympatric with 15H F 1,003* HH or hybridizing with An. freeborni at any of the 15H F 1,003 HH locations sampled in this study. However, since 18H H 1,063* HH An. herrnsidoes not have a unique X chromo- 18H H 1,063 HH 18H H 1,063 HH some,the only way to identify this speciesreli- 19 HF F 1,074* HH ably is by use of a rDNA probe or by crossing 19 F 1,074 HH studies. 19 F 1,074 HH Prior to this study, An. herrnsi was known * Collectionsof Anophelesh.ermsi from southern only as far north as Santa Maria, San Luis California.1,003: Riverside Co., Rubidoux,Carlson Obispo County (Cope et al. 1988). It is now Pk; 1,063:Ventura Co., Piru Creek;1,074: San Luis apparent that this speciesextends up the Cali- ObispoCo., Santa Margarita. fornia coast as far north as San Mateo County, and probably farther. Bailey et al. (1972),for example,collected An. freeborninear San Pablo able elementsand cytoplasmic incompatibilities, Bay and along the RussianRiver near Healds- seemto be the causeof sterility in hybrids. burg (SonomaCo.). Since both sitesare near the Resultsfrom the rDNA probe are consistent coast,it is possiblethat these mosquitoeswere with those of the crossing study in assigning actually An. herrnsi. only the population at Jasper Ridge (site 18) to An. hermsi. In general,polytene X chromosomesare use- ACKNOWLEDGMENTS ful diagnosticcharacters for distinguishing cryp- We thank David Dame,Donald Bailey, Gary tic speciesof anophelinetaxa (Kitzmiller et al. Buckingham,Christine Bennet, Stanton Cope, 1967,Kitzmiller 1977).There are, however,ex- Paul Kaiser, Deborah Dritz and Eric Daniels for ceptions to the distinctiveness of the X chro- their assistancein this project. We also thank mosomes.For example,An. stephcnsiListon and Steven Ingalls, Steven Romney, Alexander Hu- An. Laveran have farauti homosequential X bert, Sammie Dickson, Robert Brand, Kenneth chromosomes,and other speciesare known to Boyce,Mel Oldham,Glen Collett, Michael Mor- differ in the frequency polymorphic of a common stad, David Reed and Marsh Myers for their inversion(see review by Kitzmiller 1977). help in obtaining mosquitoes.This researchwas This investigation showsthat, contrary to the supportedin part by a grant from the U.S. Army description by Barr and Guptavanij (1988), the Medical R&D Command. Proiect Order X chromosome found in An. hermsi is not 85PP5854. unique to this species,since it is also found in mosquitoes throughout the foothill regions of north and central California, and in parts of REFERENCES CITED Oregonand Washington. It is uncertain whether Aitken, T. H. G. 1939.The Anophelescomplex in these foothill mosqui- (Diptera: toes should California Culicidae). Proc. Pacif. Sci. be classified as An. freeborni or a Congr. 6:463-484. new species.In Oregon and Washington, both Aitken, T. H. G. 1945. Studies on the anopheline X chromosome karyotypes are present in complex of western North America. Univ. Calif. Hardy-Weinberg equilibrium. However, in Cal- Publ. Entomol. 7:273-364. JouRNar,or rnB AnruRrcANMoseurro CoNTRoL Assocrerron Vol. 7, No. 2

Bailey,S. F., D. C. Baergand H. A. Christensen.1972. in the Anophel.esgambiae complex. Parassitologia Seasonaldistribution and behavior of California 15:l2l -128. anopheline mosquitoes.Proc. Calif. Mosq. Control Engels, W. R. 1980. Hybrid dysgenesisin Drosophila Assoc.40:92-101. and the stochasticloss hypothesis. Cold Spring Har- Baker,R. H. and J. B. Kitzmiller. 1963.Identification bor Symp.Quant. Biol. 45:561-b65. of certain anophelines by means of salivary gland French,W. L., R. H. Baker and J. B. Kitzmiller. 1962. X-chromosomes.Proc. N. J. Mosq. Exterm. Assoc. Preparationof mosquitochromosomes. Mosq. News 50:415-421. 22:377-383. Baker,R. H., W. L. Frenchand J. B. Kitzmiller. 1962. Kitzmiller, J. B. 1953.Mosquito geneticsand cytoge- Induced copulation in Anophel.esmosquitoes. Mosq. netics.Rev. Bras. Malariol. 5:286-359. News22:16-17. Kitzmiller, J. B. 1976.Genetics, cytogenetics and ev- Barr, A. R. 1980.Cytoplasmic incompatibility in nat- olution of mosquitoes.Adv. Genet.18:316-483. ural populations of a mosquito, Cul.expipiens L. Kitzmiller, J. B. 1977. Chromosomal differences Nature 283:71-72. among speciesof Anophelesmosquitoes. Mosq. Syst. Barr, A. R. and P. Guptavanij.1988. Anopheles hermsi 9:ll2-122. N. Sp., an unrecognizedAmerican speciesof the Kitzmiller, J. B. and R. H. Baker. 1963.The salivarv Anophelesmaculipennis Group (Diptera: Culicidae). chromosomesof Anophelcsfreeborni. Mosq. News Mosq. Syst. 20:352-356. 23:254-26I. Kitzmiller, Bianchi, U. 1968.Sulla tassonomiadi AnophelesIa- J. B., G. Frizzi and R. H. Baker. 1967. Evolution branchiaeed AnophelesatroparDus. Riv. parassitol. and speciation within the rnaculipennis 29:227-226. complex of the genusAnopheles. pp. 1b1-209.In: J. W. Wright and R. Pal (eds.). Collins, F. H., C. H. Porter and S. E. Cope. 1990. Geneticsof insect vectorsof disease.Elsevier Pub. Co., Comparisonof rDNA and mtDNA in the sibling New York. Lewallen, L. L. 1957. Paper chromatography studies speciesAnophel.es and A. hermsi. Am. J. freeborni of the Anophelesmaculipennis Trop. Med. Hyg. 42:417-423. complexin California (Diptera: Culicidae). Ann. Entomol. Coluzz| M. and A. Coluzzi. Soc. Am. 196g.Incroci tra popola- 50:602-606. zioni di Anophelnslabranchiae e AnopheLesatropar- Maldonado,Y. A., B. C. Nahlen, R. R. Roberto.M. uus.Parassitologia I 1:108-109. Ginsberg,E. Orellana,M. Mizrah, K. McBarron. Cope,S. E., R. J. Stoddard and A. R. Barr. 1988.The H. O. Lobeland C. C. Campbell.1990. Transmission distribution of an undescribed member of the of Plasmodium uiuax malafia in San Diego County, Anopheles maculipennis complex in California. California,1986. Am. J. Trop. Med. Hyg. 42:3-9. Proc. Calif. Mosq. Vector Control Assoc.56:130- Narang, S. K. and J. A. Seawright.1991. Genetic 134. differentiation among members of species com- Davidson, G. 1964.The five mating types of the An. plexes in anopheline mosquitoes (Diptera: Culici- gam.biaecomplex. Riv. Malariol. 13:167-183. dae). In: R. Sobti (ed.). Structural and functional Davidson,G. and R. H. Hunt. 19?3.The crossingand aspects of eukaryotic chromosomes.Springer/No- chromosomecharacteristics of a new sixth species rossa,Delhi (in press).