been achieved with the direct introduction of medium-maturing types which are tolerant of cold. Two cultivars from this source are at the pre-release stage there.
        The introduction of Precoz has alleviated the problem of differences in flowering stages, and the line has been used extensively in India. In fact, the features introduced by the use of Precoz led directly to the introduction of two new All-India nurseries: Extra Early, for plants maturing in 110 days or less, and Extra Bold, for plants bearing seeds averaging 35 mg or more.
        However, one suitable crossing variety alone does not give great scope for genetic improvement. An alternative was sought.
        West Asian material can be encouraged to flower at the same time as pilosae material by the use of artificially long days. This technique was used at ICARDA's headquarters in Syria. A system was developed whereby targeted crosses were made at ICARDA between exotic material with desirable traits and native South Asian material. The offspring were advanced by using a highland summer nursery, and plants from the third and fourth generations were sent to the national programs for testing and selection in the target areas.
        In this way, rust-resistant types have been developed for Bangladesh, where Barimasur-2 was released in 1993, and Barimasur-4 was released in 1995. In fact, Barimasur-4 has combined resistance to rust and stemphylium blight, and an upright structure which allows it to be intercropped with sugarcane.
        The same crossing techniques were then transferred to the region. A cross made in Faisalabad, Pakistan using the ICARDA technique resulted in the release of Masur-93, which has larger seeds and yields 31% more than the best local variety, and is resistant to ascochyta blight and rust diseases.
        The creation of genetic diversity through artificial mutation is often regarded as a means of speeding up the mutation process and identifying useful features. Work at the Nuclear Institute for Agriculture and Biology, Pakistan, has involved the use of harmless gamma-rays on lentil seeds. Among numerous mutations, plants have been identified which are high yielding, early maturing, short and upright, and which breed true. Some of these are being included in national trials.
        The story of South Asian lentil demonstrates the advantages of international cooperation in increasing the productivity of a crop which has historically poor yields--an especially useful exercise in this case, in a region which already produces half of the world's lentil. The collaboration between ICARDA and the national programs of South Asia has achieved this, and will continue. But the story is also one of biodiversity helping people to eat better food. There can be few better illustrations of just why we need to conserve the genetic treasure of agrobiodiversity--conserve it, and use it.

Dr William Erskine is Lentil Breeder, and Guy Manners is Science Editor/Writer at ICARDA.

outh Asia grows almost half the world's lentils, but the
productivity of the crop in the region has historically been poor. The lentils there were unusual, and lacked the basic variation that breeders need to work on to improve yields. Moreover, differences in the length of the growing season have always made it hard to incorporate genetic material from elsewhere.
        ICARDA has a world-wide mandate for the improvement of lentil. This makes sense, as the crop originated in the Center's home region. Along with various other crops, lentil was carried both east and west from its region of domestication in western Asia. Archaeological and linguistic evidence suggests that the crop reached the Indo-Ganges Plain about 4000 years ago. Today, very nearly half of the world's lentil is produced in South Asia. However, until recently, South Asian lentils were unique and there is strong evidence of a genetic bottleneck in the material that arrived there so long ago. ICARDA has been collaborating with the national programs of the region to increase the genetic diversity and improve lentil productivity.
        South Asian lentils have grey-green leaves compared with lentils elsewhere, and have short or no tendrils. For this reason, South Asian lentils have been separated as a distinct group, known as the pilosae. In addition, lentils from India, Bangladesh and Nepal differ from other lentils in a number of quantifiable features: they flower and mature early, they have small seeds and low yield, and they are short plants with the lowest pods produced closer to the ground than in other lentils. Lentils from Pakistan are technically pilosae, but have quantitative traits somewhere between those of the rest of South Asia and those of Afghanistan.
        In addition to these physical (morphological) differences, South Asian lentils show the lowest variation in character among all lentil-growing regions. This is despite the fact that the crop is grown over large areas and in a variety of environments within South Asia. All these differences point to the so-called founder effect: the lentils which were first introduced to the region were few and were not typical of those grown elsewhere. Over time, the plant has been grown and selected throughout the region, but the basic genetic makeup has remained unaltered. This seriously hampered any breeding efforts in the region, especially those aimed at increasing yield.
        Thus, the stage was set for collaboration between ICARDA, which has the mandate for lentil, and the national agricultural research systems of South Asia in a drive to improve lentil productivity by first widening the genetic base.
        The easiest way of adding new genes to a plant population is to introduce exotic plants. This was never going to be easy in lowland South Asia, since plants brought in from West Asia were only flowering when the local pilosae were maturing. However, one particular large-seeded cultivar, Precoz (originating from Argentina), was sufficiently early-maturing to be viable in the wetter parts of Pakistan, where it was released as "Manserha 89."
        In the highlands of western Pakistan (Balochistan province) the growing season is longer, and success has