Fig. 2. (A) IKONOS satellite image of abandoned Manikala glacier moraines offset by Karakorum Fault. Present-day Manikala glacier outwash is represented by the frozen stream (narrow white feature) cutting across the moraines north of the fault. The abandoned glacial channel (PV, the paleovalley of Manikala glacier) is east of the Manikala outwash. (B) Map of offset moraines (yellow, M1; orange, M2) and sample locations (circles with numbers). Thin lines outline main geomorphic features. Moraines M1 and M2E are offset —220 and 1520 m, respectively (offset restorations are given in figs. S3 and S4).

ern Tibet and the western Himalayas ought to be greater, because the normal component of throw on the main fault must be taken into account, along with slip accommodated on other active fault strands within and on the opposite side of the Gar pull-apart basin. These features have not been reported on at present.

At the 2s level, the millennial slip rate (10.7 ± 1.4 mm/year) derived from the Manikala moraines is larger than that (1 ± 6 mm/year) inferred from the latest interfer-ometric synthetic aperture radar (InSAR) study of western Tibet (2) (table S2). It is also almost three times larger than the morphochronologic rate of 4 ± 1 mm/year determined north of Bangong Lake (6), which also must be considered a minimum value because it only samples the northern of the two strands of the fault at this longitude (Fig. 1).

The slip-rate determinations at the Manikala glacial site sample many earthquake cycles. Hence, the effects of interseismic strain or postseismic relaxation on this measurement are largely restricted to the time interval since the last major earthquake and are therefore minimized. Two offsets of different ages yield a result consistent with east-

southeastwards motion of western Tibet at an average long-term rate of at least ~ 10 mm/year relative to the western Himalayas over the past 150 ky. The extent to which such motion reflects slip partitioning in the western Himalaya-Karakorum Fault system or extrusion of Tibet between the converging India and Tarim plates cannot be decided without additional kinematic constraints from the region between the Karakorum and Karakax faults (Fig. 1).

For the time being, we cannot rule out the possibility that the lack of a decadal InSAR displacement signal across the Karakorum Fault results from tropospheric effects, which, because of the strong Asian monsoon and exceptional topography, have been shown to be markedly seasonal in northern Tibet (19) and which remain poorly understood over the rest of the plateau. In this connection, the Global Positioning System (GPS) geodetic rate found by Baneijee et al. (10), although based on only one station 20 km north of the fault (Shiquanhe station), is 11 ± 4 mm/year (table S2), which is compatible with our millennial geomorphic rate and agrees with the geological rate of Lacassin et al. (7), 10 ± 3 mm/year averaged over 25 to 35 million years (Fig. 1).

Fig. 3. (A) 10Be exposure ages of blocks sampled along the M1 and M2 moraine crests (numbers refer to Fig. 2B and table S1). (B) Comparison of the distribution of sample ages (10-ky bins) with the SPECMAP 518O proxy climate curve (76) shows a simple correlation of main moraine emplacement periods with MIS 2 (the LGM), MIS 3, and the coldest parts of MIS 6. (518O increases during glacial advances as 16O is preferentially sequestered in the polar ice caps.)

If the observed disparity is real, the variation in rate must be related to fault zone mechanics and measurement interval. Deformation profiles across strike-slip faults measured geodetically can vary through the earthquake cycle depending on the rheology of the crust and mantle below the seismo-genic zone (20-22). Geodetic rates derived from standard dislocation models (2, 10) are sensitive to these variations and may underestimate slip rates, particularly if the fault is late in its earthquake cycle. Kinematic compatibility constraints and recent observations suggest that strike-slip faults that intersect, such as the Garlock/San Andreas (23) and East/North Anatolian (24), might be prone to slip-rate fluctuations over time scales longer than the typical seismic cycle. The Karakorum and Altyn Tagh faults also intersect and have not ruptured in large earthquakes since the inception of instrumental seismology; the decadal and millennial slip rates on the central Altyn Tagh fault differ by a factor of three (5, 25-27). On the basis of three-dimensional paleoseismology at Wrightwood, California, Weldon et al. (28) show that the slip rate on the San Andreas Fault varied between 8.9 and 2.4 cm/year over the past 2,000 years, spanning 14 seismic cycles. Complementary, long-term variations in slip rates appear to have existed between the San Andreas and San Jacinto faults since 1.5 million years ago (29). On the basis of emerging data, we thus conclude that secular variations in slip rate may be the rule, rather than the exception, on most faults, and that geodetic and geologic data need not be in agreement.

References and Notes

1. J. P. Avouac, P. Tapponnier, Geophys. Res. Lett. 20, 895 (1993).

2. T. J. Wright, B. Parsons, P. C. England, E. J. Fielding, Science 305, 236 (2004).

3. J. Van der Woerd et al., Geophys. J. Int. 148, 356 (2002).

4. C. Lasserre et al., J. Geophys. Res. 107, art. no. 2276 (2002).

5. A. Meriaux et al., J. Geophys. Res. 109, art. no. B06401 (2004).

6. E. T. Brown etal., J. Geophys. Res. 107, art. no. 2192 (2002).

7. R. Lacassin et al., Earth Planet. Sci. Lett. 219, 255 (2004).

8. Q. Liu, thesis, Universite Paris VII (1993).

10. P. Banerjee, R. Burgmann, Geophys. Res. Lett. 29, art. no. 1652 (2002).

11. M. A. Murphy et al., Geol. Soc. Am. Bull. 114, 428 (2002).

13. R. Armijo, P. Tapponnier, H. Tonglin, J. Geophys. Res. 94, 2787 (1989).

14. A piercing point is the intersection of a fault with linear landscape features such as abandoned stream channels, terrace risers, or morainic ridges that have been cut and offset by fault motion. We estimated the errors on the offset determinations by translating the opposing sides of the fault until an obvious visual mismatch of the relevant piercing points was observed. Because the spatial resolution of the IKONOS satellite imagery used in this study is 1 m, the resulting error estimates are dominated by the width of the features used and become larger (in absolute terms) as the features become older. The crests of M1 and M2 are 10 m and 50 m wide, respectively, at their widest in the field and on the satellite images.

15. Sampling was performed in September 2001. Samples were typically well-embedded blocks of vein quartz ~20 cm in diameter or, occasionally, chips removed from exposed parts of larger samples. Beryllium extraction procedures and production rate calculations follow those described by Meriaux et al. (5, 30). The ratios of cosmogenic 10Be to stable isotope 9Be were determined by accelerator mass spectrometry at the Lawrence Livermore National Laboratory Center for Accelerator Mass Spectrometry.

16. J. Imbrie et al., in Milankovitch and Climate, Part I, A. Berger, J. Imbrie, J. Hays, G. Kukla, B. Saltzman, Eds. (Reidel, Boston, 1984), pp. 269-305.

17. R. C. Finkel, L. A. Owen, P. L. Barnard, M. W. Caffee, Geology 31, 561 (2003).

18. L. A. Owen etal., Geol. Soc. Am. Bull. 115,1356 (2003).

19. C. Lasserre, G. Peltzer, F. Crampe, Eos 42, F271 (2001).

21. J. C. Savage, W. H. Prescott, J. Geophys. Res. 83, 3369 (1978).

22. H. Perfettini, J. P. Avouac, J. Geophys. Res. 109, art. no. B06402 (2004).

23. G. Peltzer, F. Crampe, S. Hensley, P. Rosen, Geology 29, 975 (2001).

24. A. Hubert-Ferrari etal., Geophys. J. Int. 153,111 (2003).

25. R. Bendick, R. Bilham, J. Freymueller, K. Larson, G. H. Yin, Nature 404, 69 (2000).

26. K. Wallace, G. H. Yin, R. Bilham, Geophys. Res. Lett.

28. R. Weldon, K. Scharer, T. Fumal, G. Biasi, GSA Today 14, 4 (2004).

29. R. A. Bennett, A. M. Friedrich, K. P. Furlong, Geology

30. Materials and methods are available as supporting material on Science Online.

31. This work was performed under the auspices of the

U.S. Department of Energy by University of California Lawrence Livermore National Laboratory under contract W-7405-Eng-48 under the sponsorship of the Laboratory Directed Research and Development program (report no. UCRL-JRNL-206541). Also supported by the Institut National des Sciences de l'Univers, Centre National de la Recherche Scientifique (Paris, France), through programs Imagerie et Dynamique de la Lithosphère and Interieur de la Terre, and by the China Earthquake Administration and the Ministry of Lands and Resources (Beijing, China).

Supporting Online Material DC1

Materials and Methods Figs. S1 to S5 Tables S1 to S2 References and Notes

21 September 2004; accepted 15 December 2004 10.1126/science.1105466

Speciation by Distance in a Ring Species

Darren E. Irwin,1* Staffan Bensch,2 Jessica H. Irwin,1 Trevor D. Price3

Ring species, which consist of two reproductively isolated forms connected by a chain of intergrading populations, have often been described as examples of speciation despite gene flow between populations, but this has never been demonstrated. We used amplified fragment length polymorphism (AFLP) markers to study gene flow in greenish warblers (Phylloscopus trochiloides). These genetic markers show distinct differences between two reproductively isolated forms but gradual change through the ring connecting these forms. These findings provide the strongest evidence yet for "speciation by force of distance'' in the face of ongoing gene flow.

Traditional models emphasize geographic separation as a necessary prerequisite to speciation (1, 2). Although experiments and theory indicate that species can form despite ongoing gene flow (3-5), there are very few known examples in nature (2). Some studies have demonstrated divergence despite gene flow (6, 7), but they do not enable an assessment of reproductive isolation because the divergent forms remain geographically separated. Species are usually defined as groups of interbreeding populations reproductively isolated from other such groups (1, 2), and this can only be critically examined if different populations regularly come into contact in nature.

There are a few examples where repro-ductively isolated populations coexist while being connected by apparently gradual variation around geographic barriers ["ring

''Department of Zoology, University of British Columbia, 6270 University Boulevard, Vancouver, BC, Canada V6T 1Z4. 2Department of Animal Ecology, Lund University, S-223 62 Lund, Sweden. 3Depart-ment of Ecology and Evolution, University of Chicago, 1101 E. 57th Street, Chicago, IL 60637, USA.

*To whom correspondence should be addressed. E-mail: [email protected]

species"; reviewed in (8)]. In theory, ring species enable us to trace the process by which one species diverges into two. They also potentially show that reproductive isolation can arise in the face of gene flow (1, 8-10). However, a clear pattern of a gradual genetic variation has not previously been observed in a ring species. Here, we use molecular markers to show that two reproductively isolated forms of greenish warbler (Phy/loscopus trochi/oides) are connected by gene flow through a ring of populations, providing the strongest empirical evidence yet for "speciation by force of distance" (1, 9).

Two forms of greenish warbler, one in west Siberia (P. t. viridanus) and one in east Siberia (P. t. pMmbeitarsus; Fig. 1), coexist without interbreeding in central Siberia and can therefore be considered separate species (10). These forms are connected by a chain of populations to the south that encircles the high-altitude desert of the Tibetan Plateau, which is not inhabited by the warblers. Through this chain of populations, traits such as color patterns, morphology, and behaviors (song and song recognition), change gradually, demonstrating a smooth gradient in forms between two species (10, 11). There is evidence that all of these traits are under selection in the Phy/loscopus warblers (10-15); it is therefore unclear that such traits can be used to infer gene flow. To directly measure genome-wide genetic relationships, we used amplified fragment length polymorphism (AFLP) markers (16).

From 105 greenish warblers at 26 sites throughout the breeding range we obtained 62 AFLP markers that were variable and could be scored unambiguously as present or absent in each individual (17). West Siberian viridanus and east Siberian p/umbeitarsus are clearly separated in AFLP genotypes, which confirms that the two taxa are genetically distinct. In contrast, AFLP genotypes change gradually through the ring of populations to the south (Fig. 1). The genetic gradient in the AFLP genotypes around the southern ring of populations is best seen in a plot of pairwise AFLP distances versus pairwise geographic distance (Fig. 2). Geographic distances were measured under the assumption that no genes flow across the uninhabited area in the center of the ring or between viridanus and p/umbeitarsus in central Siberia. Thus, "corrected" distances between west Siberian (viridanus) and east Siberian (p/umbeitarsus) populations were measured through the long chain of populations running to the south of Tibet, through the Himalayas. Genetic distance and corrected geographic distance are strongly correlated (Mantel's r = 0.782, P = 0.0003), consistent with a pattern of isolation by distance (18) around the ring. An alternative analysis based on pairwise FST distances between populations produces similar results (Mantel's r = 0.677, P = 0.0012; table S1 and fig. S3).

On the basis of these results, we conclude that there is no break in gene flow through the ring of populations, except between the divergent forms viridanus and p/umbeitarsus in central Siberia. Thus all populations have been recently connected by at least some gene

Supplements For Diabetics

Supplements For Diabetics

All you need is a proper diet of fresh fruits and vegetables and get plenty of exercise and you'll be fine. Ever heard those words from your doctor? If that's all heshe recommends then you're missing out an important ingredient for health that he's not telling you. Fact is that you can adhere to the strictest diet, watch everything you eat and get the exercise of amarathon runner and still come down with diabetic complications. Diet, exercise and standard drug treatments simply aren't enough to help keep your diabetes under control.

Get My Free Ebook

Post a comment