all, Eds. (Elsevier, New York, 1985), pp. 203-243. 4. We include the Nubian Desert of Sudan, the Eastern and Western deserts of Empt, the Libyan Desert, and all other subdeserts of North Africa contiguous to the Sahara in our figures for the Sahara. 5. D. W. Goodall and R. A. Perry, Eds., Arid-Land Ecosystems: Structure, Functioning, and Management (Cambridge Univ. Press, Cambridge, 1979), vol. 1; W. G. McGinnies, ibid., pp. 299-314; A. Shmida, in Ecosystem ofthe World: Hot Deserts and Arid Shmb- lands, M. Evenari, I. Noy-Meir, D. W. Goodall, Eds. (Elsevier, New York, 1985), pp. 23-77. 6. M. Evenari, in Ecosystem of the World: Hot Desem ard Arid Shrublands, M. Evenari, I. Noy-Meir, D. W. Good- all, Eds. (Elsevier, New York, 1985), pp. 1-22. 7. F. R. Cana, Geogr. J. 46, 333 (1915); E. W. Bovill, J. Afr. Soc. 20, 175 (1921); ibid., p. 259; E. P. Stebbing, Geogr. J. 85, 506 (1935); F. Rodd, ibid. 91, 354 (1938); E. P. Stebbing, ibid., p. 356; A. Aubreville et al., Bois For. Trop. 148, 3 (1973). 8. E. Eckholm and L. R. Brown, Worldwatch Paper 13 (Worldwatch Institute, Washington, DC, 1977), p. 1. 9. J. Smolowe, Time 127 (no. 33), 36 (1987). 10. C. Norman, Science 235, 963 (1987). 11. W. S. Ellis, Natl. Geogr. 172, 140 (1987). 12. H. F. Lamprey, "Report on the desert encroachment reconnaissance in Northern Sudan" (Unesco/ UNEP) [Deserf$cation Control Bull. 17, 1 (1988)l. 13. S. E. Smith, J. Soil Water Comerv. 41, 297 (1986). 14. M. Kassas, in Arid Lands in Tramition, H. E. Dregne, Ed. (AAAS, Washington, DC, 1970), pp. 123-142. 15. H. N. Le Houerou,Ann. Alger. Geogr. 6 , 2 (1968). 16. S. E. Nicholson, Weather 44, 47 (1989). 17. P. J. Lamb, R. A. Peppler, S. Hastenrath, Nature 322, 238 (1986). 18. C. Toupet, C . R . Somm. Seances Soc. Biogeogr. 48, 39 (1972). 19. U. Hellden, Drought Impact Monitoring (Lund Uni- versity Naturgeogratiska Institute, Lund, Sweden, 1984). 20. A. R. Malo and S. N. Nicholson, J. Arid Environ. 19, 24 (1990). 21. P. J. Curran, Philos.Trans. R. Soc. London Ser. A 309, 257 (1983); C. J. Tucker and P. J. Sellers, Int. J. Rentote Seming 7, 1395 (1986); P. J. Sellers, ibid. 6 , 1335 (1985); J. L. Monteith, Philos.Trans. R. Soc. London Ser. B 281,277 (1977); J. L. Hatfield, G. Asrar, E. T. Kanemasu, Rentote Seming Environ. 14, 65 (1984); C. S. T. Daughtry, K. P. Gallo, M. E. Bauer, Agron. J. 75, 527 (1983); G. Asrar, E. T. Kanemasu, G. P. Miller, R. L. Weiser, IEEE Trans. Geosci. Remote Sensing GE24, 76 (1986). 22. C. J. Tucker, B. N. Holben, J. H. Elgin, J. E. McMurtrey, RentoteSensingEnviron. 11, 171 (1981); G. Asrar, E. T. Kanemasu, R. D. Jackson, P. J. Pinter, ibid. 17,211 (1985); C. J. Tucker, C. L. Vanpraet, M. J. Sharman, G. Van Ittersum, ibid. 17, 233 (1985); G. Gosse et al., Agronomie 6, 47 (1986); S. D. Prince, Int. J. Remote Sensing 12, 1301 (1991). 23. Precipitation varies from <lo0 mm in the Sahara Desert to -400 mm at the southern boundary of the Sahel zone; a precipitation gradient of -1 mm km-' occurs from -18"N to -1O0N (2, 24). The Sahara Desert, Sahel zone, and their transition zone have been defined as having the following rainfall values: 0 to 100 mm year-', Sahara Desert; 100 to 200 mm year-', Saharan- Sahelian transition zone; and 200 to 400 mm year ', Sahel proper. Rainfall in this area is unimo- dal and occurs from July to October; the rest of the year is dry (2, 3, 24, 25). 24. F. White, The Vegetation of Africa (Unesco, Paris, 1983). 25. H. Breman and C. T. de Wit, Science 221, 1341 (1983). 26. Data from NOAA-6 (1980), NOAA-7 (1981 to 1984), NOAA-9 (1985 to 1988), and NOAA-11 (1989 to 1990) were used from 1 July to 31 October of 1980 to 1990. Daily AVHRR 4-km data were acquired for Africa, the NDVI was formed from channel 1 and channel 2 as (2 - 1)/(2 + l ) , a cloud mask using channel 5 was applied labeling everyhng colder than 12°C as cloud, and the data were mapped to an equal-area projection with a -7.6-km grid cell. The daily mapped data were formed into 10-day images and averaged for each year. Formation of composite images minimizes atmospheric effects, scan angle effects, cloud con- tamination, and solar zenith angle effects [B. N. Holben, Int. J. Remote Sensing 7, 1417 (1986); M. Shibayama, C. L. Wiegand, A. J. Richardson, ibid., p. 233; K. J. Ranson and C. S. T. Daughtry, IEEE Tram. Geosci. Remote Sensing GE25, 502 (1987)J. Calibration corrections after Y. J. Kaufman and B. N. Holben (Int. J. Remote Sensing, in press) were applied. A total of 4500 orbits of AVHRR data were used. 27. A table with the estimated 200 mm year-' isoline boundary from 15.5"W to 38.5"E at 0.5" longitude increments from 1980 to 1990 is available from th authors upon request. 28. H. N. Le Houerou and C. H. Hoste, J. Range Manan. 30. 181 (19771; M. C. Rutherford..S. Afr. J. sct: 76, 53 (1980); M. K. Seely, ibid. 74, y95 11978). 29. w e Lank D. Rosenfelder, J. Rosenfelder, and R. Rank for assistance. 12 February 1991; accepted 15 May 1991 Organic Molecular Soft Ferromagnetism in a Fullerene C6() The properties of an organic molecular ferromagnet [C,,TDAE,,,,; TDAE is tetra- kis(dimethylamino)ethylene] with a Curie temperature T, = 16.1 kelvin are described. The ferromagnetic state shows no remanence, and the temperature de magnetization below T, does not follow the behavior expected of a conv ferromagnet. These results are interpreted as a reflection of a three-di leading to a soft ferromagnet. T HE QUESTFOR A NONPOLYMERIC organic ferromagnet has intensified during the past 5 years with varying results (1-10). Curie temperatures (T,'s) on the order of 1 to 2 K have been observed (7-12). An organometallic molecular ferro- magnet with Tc of -4 K has been known for sometime (13), and similar systems with T, of 6.2 K (14) and 8.8 K (15) have recently been reported. Polymeric ferromag- nets with TCys claimed to be greater than 300 K have also been described but their characterization remains incomplete (2, 16), and the possibility of a polaronic, polymeric ferromagnet has appeared (17). We report on the preparation and preliminary charac- terization of an organic molecular solid with a transition at 16 K to a soft ferromagnet state. Our interest in the preparation of materi- als based on the reduction (n-doping) (18, 19) of fullerenes (20-25) prompted us to explore the use of strong organic reducing agents such as tetrakis(dimethy1amino) ethylene (TDAE). Addition of a 20 M ex- cess of the donor to a solution of C,, in toluene in a dry box afforded a black micro- crystalline precipitate of C,,(TDAE),,,, [or (C,,) ,, ,,TDAE] (26). The extremely air- sensitive (27) solid was washed with toluene and loaded into a capillary tube prepared from a Pasteur pipette, which had been previously sealed at the narrow end. The wide bore end of the pipette was then connected to a 5-mm vacuum stopcock, and the pipette was removed from the dry box to seal the sample under He in a vacuum line at -2 cm abovethe powder fill line. The sample was then cooled in zero field in an ac susceptometer and then allowed to warm in an applied field of 50.1 Oe. Surprisingly, the transition observed was apparently to a ferromagnetic state. The sample was studied in greater detail by dc magnetization (M) measurements tak- en on a Quantum Design superconducting quantum interference device (SQUID) magnetometer. Care was taken to ensure that the sample was nor exposed to a field gradient greater than 0.03% of the applied field. In Fig. 1 we show M as a function of T for the sample cooled and warmed in an applied field Ha - 1 Oe. Two significant observations can be made from Fig. 1: (i) although M increases sharply below Tc - 16.1 K as expected for a ferromagnet, the temperature dependence of M does not fol- low-that of conkntional mean field theory; P.-M. Allemand, K. C. Khemani, A. Koch, F. Wudl, Deoartments of Chemistrv and Phvsics and Institute for and (ii) within there is ~ o i ~ m e r s and Organic ~dlids, ~dversity of California, no hysteresis between cooling and warming. Santa Barbara, CA 93106. K. Holcwr, S. Donovan, G. Griiner, Department of The Structure in M(T) lo is Physics, University of California, Los Angeles, CA dependent; similar measurements in applied 90024. J. D. Thompson, Los Alamos National Laboratory, Los fieids of 10, 100, and 1000 Oe showthat Alamos, NM 87545. with increasing field the minimum at 8 K 19 JULY 1991 REPORTS 301