Sie sind auf Seite 1von 3

experiments and the univariate mapping 6. G. E. P. Box and N. R.

Draper, Evolutionary Op- case of hybridization in which one F, and


eration (Wiley, New York, 1969).
allows not only the determination of the 7. A simplex is a geometric figure defined by a num- three backcross hybrid individuals have
best reaction conditions but also the speci- ber of points (n + 1) equal to one more than the been unequivocally recognized in nature.
number of dimensions (n) of the factor space. This
fication of tolerances (1) within which the term does not have the same meaning as the term Drosophila setosimentum and D. ochro-
factors must be held to achieve this high "simplex" used in linear programming [see, for ex- basis are near-sibling species endemic to
ample, G. B. Dantzig, Linear Programming and
yield. In particular, the sensitivity of reac- Extensions (Princeton Univ. Press, Princeton, the island of Hawaii (5). Drosophila setosi-
N.J., 1963)].
tion yield to T2 suggests to us the possi- 8. W. Spendley, G. R. Hext, F. R. Himsworth, Tech- mentum occurs in rain forests between 600
bility that other reactions might also ex- nometrics 4, 441 (1962). and 1600 m altitude. At five widely sepa-
9. J. A. Nelder and R. Mead, Comput. J. 7, 308
hibit a marked temperature optimum (1965). rated locations, all above 1000 m, it is ac-
which does not necessarily coincide with 10. Given n + I simplex vertices represented as n-di- companied by D. ochrobasis. Both species
mensional vectors W, N, ..., B ranked in order of
the boiling point of the solvent (a choice response Sw (worst), SN., SB (best) and cen- are rare and local, being largely confined
commonly employed for reasons of conve- troid P = (N + ... + B), eliminate W and evalu- to kipukas (islands of vegetation isolated
ate the response SR at R = P + (P W). If
nience). A more detailed study of this and SN 5-SR < SB, retain R. If SR > SpB, evaluate SE at in the midst of newer lava flows) or col-
other reactions (including catalytic sys- E = P + 2(P W): if SE > Se, retain E; if SE <
-
lapsed lava tubes where their rare host
SB, retain R. If SR < SN: if SR > SW, evaluate Sc,
tems), investigating the effects of more at and retain Cr = P + '(P - W); if SR < Sw, eval- plants grow. Although no reliable means
uate Scw at and retain Cw = P (P W). If Sc,
than two factors, is in progress. orScw < SN, set Sw = SN, W - N, SN = SC, or exists for the separation of females on
WALTER K. DEAN Sc. and N = Cr or Cw. See C. W. Lowe [Trans. morphological grounds, males of the two
Inst. Chem. Eng. 42, T334 (1964)] for an example
KATHARINE J. HEALD of a calculation work sheet. species are easily distinguished by a sec-
Department ofChemistry, 11. S. N. Deming and S. L. Morgan, Anal. Chem. 45, ondary sexual difference in wing pattern.
278A (1973).
Emory University, 12. W. K. Dean and P. M. Treichel, J. Organometal. Wild-caught females were separated
Atlanta, Georgia 30322 Chem. 66,87 (1974).
13. Reactions were carried out as follows: 15 ml of from males at capture. In the laboratory,
STANLEY N. DEMING THF and 4.40 ml of a 0.50M solution of C,H,Na
in THF were added to 2.00 mmole (0.528 g) of
each female was placed in a separate cul-
Department of Chemistry,

Downloaded from http://science.sciencemag.org/ on February 8, 2019


Mo(CO)6. The mixture was refluxed (650C) tor t. ture vial and allowed to produce offspring.
University of Houston, and then cooled to T2; 4.40 ml of 0.50M Such a culture (an isofemale line) usually
CICSN(CH,)2 in THF were then added, and the
Houston, Texas 77004 mixture was stirred for 22 hours in a water bath yields F, males conforming exclusively to
held at T2. The reaction yield was determined by either one species or the other. From each
References and Notes adding 100 ml of THF and a weighed sample of
benzonitrile (about 3.0 g) to the reaction mixture; isoline, seven (sometimies fewer) F, larvae
1. A. L. Wilson, Talanta 17, 21 (1970). the infrared spectrum of the solution was deter- were taken, and an :acetoorcein squash
2. M. Friedman and L. J. Savage, in Techniques of mined on a spectrophotometer (Perkin-Elmer
Statistical Analysis, C. Eisenhart, M. W. Hastay, model 467), and the height of the absorption maxi- preparation of the salivary gland chromo-
W. A. Wallis, Eds. (McGraw-Hill, New York, mum of the product (1841 cm-') was compared to
1947), chap. 13. that of benzonitrile (2245 cm-'). somes from each was prepared. From each
3. J. J. Carroll, N. Smith, A. L. Babson, Biochem. 14. This research was supported by a university re- smear the banding order of all five major
Med. 4, 171 (1970); C. C. Allain, L. S. Poon, C. S. search grant from the Emory University Research
G. Chan, W. Richmond, P. C. Fu, Clin. Chem. 20, Fund and by the National Science Foundation chromosome arms (X, 2, 3, 4, and 5) was
470 (1974). (grant GP-3291 1). read directly along the full length of the
4. G. E. P. Box, Biometrics 10, 16 (1954).
5. S. L. Morgan and S. N. Deming, Anal. Chem. 46,. chromosome, with the use of spaced land-
1170 (1974). 14 April 1975 mark areas to confirm the sequences. The
banding orders of all paired homologs
were thereby determined. Inverted sections
were read in either homozygous or hetero-
Drosophila Hybrids in Nature: zygous state. The sequence of the hem-
Proof of Gene Exchange Between Sympatric Species izygous X in each male larva was also de-
termined.
Abstract. Genetic studies of two closely related endemic Hawaiian species show that in These cytological data permit inferences
one area of sympatry about 2 percent of the naturally occurring individuals are hy- about the state of the natural population
brids. More than 20 times this many would be expected if the population consisted of a from which the wild flies were drawn. Ex-
single panmictic unit. Despite hybridization, natural selection appears to maintain the es- cept in the Kahuku Ranch population, to
sential integrity of each separate gene pool. be described later, all the F, test larvae
(195 female and f45 male larvae) exam-
In the Kahuku area, near the south end males. In D. pseudoobscura and D. per- ined from 56 wild D. ochrobasis females
of the island of Hawaii, natural inter- similis, close genetic scrutiny of the proge- have shown a uniform sequence in chromo-
specific hybridization has been detected be- ny of 27,099 wild-caught flies from sym- some X (ochrobasis standard X). In addi-
tween a pair of sympatric endemic species. patric areas yielded four cases of F, hy- tion, all larvae are homozygous for a
This circumstance is very unusual for the bridization (3). In each case, a wild female chromosome 2 inversion (2k), and all lack
genus, the other species of which tend to of one species had mated with a male of the an inversion (21) near the opposite end of
have isolated gene pools which coexist in other species. Considerable doubt exists, the same chromosome. This latter inver-
nature without evidence of gene exchange however, as to whether the interspecific sion is fixed in all but a North Kona, Ha-
(1). Nevertheless, laboratory crosses be- matings observed actually took place in waii, population of D. setosimentum. In
tween sympatric species are frequently nature; since the wild flies were not sepa- addition, the commonest gene order of
possible and sometimes one or both sexes rated at capture, the interspecific mating chromosome 4 of D. ochrobasis differs
of F, hybrids are fertile (2). Accordingly, if could have occurred in the collecting vials from the standard 4th chromosome of D.
reproductive isolation became weak, the on the way to the laboratory. Only a single setosimentum by six inversions. These are
potential for gene exchange would exist. instance of a backcross hybrid was detect- spread in a roughly tandem manner over
In an extensive literature on Drosophila ed. In D. metzii and D. pellewae, in- the length of the chromosome.
species extending over 50 years, only two trogression in nature may be strongly in- Similarly, 154 female and 164 male test
cases suggest the possibility of in- ferred from a face color polymorphism, larvae have been recognized from sympat-
trogression in nature. In both, laboratory but critical genetic proof of hybridization ric areas as the progeny of 54 wild D. se-
crosses give fertile F, females and sterile F, in nature is lacking (4). We report here a losimentum flies. In allopatric areas, 221
806 SCIENCE, VOL. 189
D. setosimentum were also examined The second case was that of a larva col- going free genetic recombination. Accord-
(1062 test larvae). Collections from Ka- lected from a natural oviposition site, a ingly, we hold that the hybridization cir-
huku Ranch are excluded from these fermenting branch of the plant Cler- cumstance is a manifestation of recent (in
counts. montia. Fourteen third instar larvae from the geological sense) divergence between
Accordingly, from two to nine chromo- the branch were smeared and the tissues of these species. Although each species has
somal markers serve to distinguish each in- the same larvae were subjected to electro- apparently evolved a recognizable and in-
dividual of each species. Their homo- phoresis. In addition, eight emerging tegrated gene pool of its own, each remains
zygosity and their distribution in the ge- adults from the same branch were sub- open to the acquisition of small amounts of
nome is such that three of the five major jected to electrophoresis. Of these 22 speci-genetic variability from the other.
chromosome pairs are well marked. The mens, 20 were D. ochrobasis, one was D. This leaky boundary merely represents a
lack of heterozygotes in samples of the size setosimentum, and one was a female back- stage in what may be a rather common
reported above means that the probability cross hybrid. The latter displayed the pattern of allopatric speciation in general.
of incorrect diagnosis of species and of double polymorphism in chromosome 2 Two separated populations may come
failing to recognize F, and backcross hy- but was homokaryotypic for the D. ochro- back into contact after each has virtually
brids between the species is very small. basis X and 4th chromosomes. Finally, completed the integration of its new gene
Had we been dealing with one population two wild male flies having D. setosimen- pool. If there is only weak reproductive
at Hardy-Weinberg equilibrium, 55 het- tum wing patterns, out of 40 electrophoret- isolation full recombination might occur.
erozygotes at any one of the autosomal ically tested but not studied cytologically, In this, as in many other cases in plants
markers would have been expected. proved to be heterozygous for the highly and animals, the reciprocal flow of genes
In the laboratory, hybrids between the distinctive "null" allele of D. ochrobasis appears to be insignificant probably be-
species are fairly easily made. The resul- at the Est- 1 locus. cause of natural selection working against
tant chromosomal heterozygotes are those The foregoing data constitute proof that inferior F2 and backcross combinations.

Downloaded from http://science.sciencemag.org/ on February 8, 2019


that would be expected from the readings gene exchange actively occurs between The extent of such hybrid zones between
of the two sets of homozygous states. Thus, contemporary populations of these species closely related species varies greatly; thus
the D. ochrobasis X chromosome in fe- the broad zones found in some plants differ
in this locality. Of 180 specimens studied, 4
male hybrids is observed in heterozygous (2.2 percent) were hybrids. If the popu- greatly from the small zone found in this
state with one of four alternate gene orders lation were at Hardy-Weinberg equilibri- Drosophila case.
known to be polymorphic within D. setosi- um, 89 heterozygotes, more than 20 times A recent theory suggests that a species,
mentum. In addition, the expected two-in- the observed number, would be expected at as distinguished from an infraspecific pop-
version difference (2k 21+/2k+ 21) appears the autosomal 2k inversion region alone. ulation, is characterized by a unique in-
in chromosome 2 and the expected six-in- The sites at which collections were made tegrated internal system of genetic balance
version heterozygote appears in chromo- are in almost wholly undisturbed rain for- (10). So long as closed and balanced genet-
some 4. Further, D. ochrobasis adults have ic systems are not destroyed by recombina-
est, so that one is not tempted to invoke the
a distinctive fixed electrophoretic differ- idea that man has "hybridized the habitat" tion, populations can coexist even in the
ence relative to D. setosimentum. This (7) and thus provided unusual ecological presence of considerable interbreeding.
may be observed at a ,B-naphthyl acetate conditions which might have been con- The case we discuss conforms to this view.
esterase (Est- I) locus (6). ducive to the hybridization process or to This is to be expected on a large isolated
Cytological or electrophoretic assays (or the biological success of its genetic prod- tropical island which is geologically new
both) have been carried out on 180 wild ucts. and which has rain forests which continue
specimens collected from the Kahuku Discovery of F and backcross hybrids is to be dissected by recent lava flows.
Ranch, Kau District, island of Hawaii. of particular interest in view of Carson and H. L. CARSON
Polytene chromosomes of seven F, larvae Johnson's suggestion (5) that these species Department of Genetics,
from each isofemale line were scored, and have hybridized in the past, as evidenced University of Hawaii, Honolulu 96822
each larval corpse was subjected to elec- by the existence of a series of peculiar P. S. NAIR
trophoresis. Parallel cytological and elec- kinked homologs ("complex chromo- Department of Biology, Southern
trophoretic data was thus provided for a somes") which have been found segregat- Illinois University, Edwardsville 62025
total of 744 test larvae. Most wild-caught ing within most sympatric populations of F. M. SENE
males were subjected to electrophoresis both species. In gene order these kinked Department ofBiology CP1 1461,
without cytological examination, but some chromosomes resemble certain chromo- University ofSao Paulo,
were analyzed by crossing to laboratory somes of the other species; this appears Sao Paulo, 05421 Brazil
virgins from a monomorphic stock and to have resulted from reciprocal exchange
subjected to electrophoresis after F, larvae occurring in past generations. Why they References and Notes
had been produced by their mates. have become complexly kinked is not 1. J.Genus T. Patterson and W. S. Stone, Evolution in the
Drosophila (Macmillan, New York, 1952).
Four exceptional individuals were known. 2. E. Craddock, Evolution 28, 593 (1974); H. Carson,
found. The first was a wild female that Geographical speciation theory can be 3. Th. ibid. 8, 148 (1954).
Dobzhansky, Am. Nat. 107, 312 (1973).
transmitted a D. setosimentum X chromo- invoked to explain the origin of these spe- 4. S. Pipkin, Evolution 22, 140 (1968).
some to her sons but showed a D. ochro- 5. H. Carson
cies in an ecosystem that is continually 6. P. Nair, F. and W. Johnson, ibid. 29. 11 (1975).
Sene, H. Carson, Genetics 80, s60
basis-D. setosimentum heterozygous X in being fractionated by lava flows (8). Nev- (1975).
her female progeny. Autosomes 2 and 4 of ertheless, the conventionally isolated gene 8.7. E.H. Anderson, Evolution 2, 1 (1948).
Carson, Stadler Symposia 3, 51 (1971).
both sexes showed the hybrid condition in pools of each are not now closed to ex- 9. V.NewGrant, Plant Speciation (Columbia Univ. Press,
York, 1971).
all test larvae. As the female was isolated change with the other at places where they 10. H. Carson, Am. IV~at. 109, 83 (1975).
at capture, this is an unequivocal case come in contact. As in many plants, in- 11. Supported by NSF grants GB27586 and GB29288
to the University of Hawaii and by Office of Re-
(based on heterozygosity for four indepen- trogression (9) is occurring and apparently search and Projects, Southern Illinois University.
has occurred in the past. Despite this, how- F.M.S. is postdoctoral fellow of Fund. Amparo a
dent chromosomal marke") of a wild D. Pesquiisa do Est, S. Paulo, Brasil. We thank K. Y.
setosimentum femalck hich had mated ever, the gene pools of the two species give Kane;hiro and L. T. Teramato for help.
with a D. ochrobasis rdale Wnature. no evidence of flowing together and under- I May 1975; revised 3 June 1975
5 SEPTEMBER 1975 807
Drosophila hybrids in nature: proof of gene exchange between sympatric species
HL Carson, PS Nair and FM Sene

Science 189 (4205), 806-807.


DOI: 10.1126/science.1162353

Downloaded from http://science.sciencemag.org/ on February 8, 2019


ARTICLE TOOLS http://science.sciencemag.org/content/189/4205/806

REFERENCES This article cites 7 articles, 0 of which you can access for free
http://science.sciencemag.org/content/189/4205/806#BIBL

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Use of this article is subject to the Terms of Service

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the
Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. 2017 © The Authors, some rights
reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S.
Government Works. The title Science is a registered trademark of AAAS.

Das könnte Ihnen auch gefallen