ABSTRACT Smallholders often cannot obtain the agroforestry germplasm they require. It has been suggested that commercial smallholder production of germplasm could help resolve this problem, while securing benefits such as genetic conservation and income generation. We evaluate four such initiatives. We found that worthwhile income was generated, and we suggest ways that returns could be increased. We also found that the initiatives had contributed to genetic conservation, although we argue that such outcomes may be more elusive for timber trees than for non-destructively harvested tree crops. The four initiatives fell short of the “win-win” ideal of germplasm production, i.e. that of smallholders profiting through sale of livelihood-enhancing germplasm to other smallholders, principally because sales tend to be to industrial or institutional clients rather than to smallholders. We suggest that, for the potential of smallholder germplasm production to be fully evaluated, detailed characterization of farmer demand and willingness to pay for agroforestry germplasm is needed. Key words: mahoganynon-timber forest productsseed pricingseed supply systemstree domesticationtree improvement

Cedrela odorata (Spanish cedar) is a neotropical broadleaf tree species that is in high demand for furniture and interior fittings. In 1998, seed collections were made from Spanish cedar in the Yucatan Peninsula, Mexico, for genetic conservation and tree improvement projects. Progeny from these collections were established in genetic trials at Bacalar, Noh Bec, and Zoh Laguna in the Yucatan. Survival at 2 years was 73.5% at Bacalar, 72% at Noh Bec, but only 20% after 6 months for several replicates at Zoh Laguna. Mean 2-year height, number of years of Hypsipyla attack on apical shoots (over 2 or 3 years) and total branching over 2 years, were 1.3m, 0.04 years, and 2.5 branches at Bacalar, and 1.9m, 0.35 years, and 1.6 branches at Noh Bec. Selection of the three best provenances (Escárcega, Bacalar, and Calakmul) would result in an estimated cross-site gain of 8.6% in 2-year height. Within-provenance, cross-site heritabilities for 2-year height, shoot borer attack, and branching were 0.1±0.02, 0.0±0.02, and 0.0±0.02, respectively. Single-site heritabilities for height at 2 years, shoot borer attack, and branching were 0.09±0.05, 0.02±0.04, and 0.09±0.05 at Bacalar and 0.16±0.07, 0.0±0.06, and 0.0±0.07 at Noh Bec. Given the known advantage of rapid growth to avoid Hypsipyla damage over the long term and the stable performance of provenances across sites for growth traits, we recommend provenance selection over multiple trials to reduce the risk of serious damage by the insect. Good provenances and other suitable seed sources, as well as proper site selection and silvicultural practices, are important tools for increasing yields from plantations of C. odorata.

Stocks of the valuable big-leaf mahogany (Swietenia macrophylla King) are declining, and trials for growth and pest resistance are needed to select material for plantations. Seeds were collected from 67 open-pollinated trees from five provenances in the Yucatan Peninsula of Mexico and planted in three provenance/progeny trials in the state of Quintana Roo, Mexico, in order to characterize genetic variation in growth traits and for Hypsipyla resistance or recovery, and to assess the potential for genetic gain. Differentiation among provenances was found only for relative height growth rate (RHGR). The total years of apical attack by Hypsipyla grandella varied by a magnitude of 100% among families but showed little heritability. After 4 years, mean height per family ranged from 328 to 564cm, 160 to 381cm, and 253 to 390cm at each site. Although heritabilities for height were too low for cross-site selection, sufficient heritability (hI2=0.26), additive genetic coefficients of variation (AGCV=22%), and type B genetic correlations (rb(f)=0.74) for RHGR across the two sites with poorer growth indicated that this trait might be used as a surrogate. This would yield an estimated gain of about 17% for the best 15% of trees. At the site with better growth, there was sufficient heritability (hI2=0.31) and AGCV (20%) for height at year 5 to obtain an estimated gain of 15% for the best 15% of trees. We suggest a selection strategy using the best germplasm from the best performing trial to exploit the resources on high quality sites, and the best material from the poorer sites for lower quality areas.

Bareroot hardwood seedling production involves intensive soil management. To increase soil organic matter (OM), nurseries commonly grow a cover crop for 1 year after every 1–2 year of seedling production. Raising soil OM levels can also be achieved through addition of soil amendments. We studied the influence of chicken manure (CM) and composted leaf, tree, and lawn trimmings (Cp) on soil properties and morphology of green ash (Fraxinus pennsylvanica Marsh.) and northern red oak (Quercus rubra L.) seedlings. CM was applied at 725, 1450, or 2900 kg ha−1 (CM725, CM1450, and CM2900, respectively) and Cp was applied at 200 m3 ha−1. Addition of CM and Cp significantly raised soil OM levels and altered soil chemical properties compared to the control (Ctrl). Root-collar diameter increased with addition of CM1450, CM2900, or Cp compared to CM725 or Ctrl plots for northern red oak, but was largest in soils amended with CM2900 for green ash. Conversely, height was greatest with addition of CM725 for northern red oak, but green ash seedlings were shorter in Ctrl plots than in all amendments except for CM725. Root volume of green ash and northern red oak seedlings was positively influenced by addition of CM or Cp. Seedling responses to nursery soil amendments vary with different forms and amounts of OM. Benefits to seedling growth through application of appropriate materials in the proper balance can improve seedling morphological quality and positively influence soil chemical properties.

Contiene: Capitulo 1: Introduccion. Capitulo 2: le indicara la forma de hacer un plan de reforestacion, para que usted sepa escoger los sitios mas adecuados de su chacra y las especies de arboles que mas se le adaptan. Capitulo 3: contiene lo que usted necesita saber sobre semillas, plantones y viveros. Capitulo 4: le dara consejos practicos sobre el modo de establecer una buena plantacion. Capitulo 5: le dira como debe cuidar o tener su plantacion; hay en el consejos sobre el deshierbe, la fertilizacion y la poda. Capitulo 6: da informacion sobre el raleo, que es un aspecto muy importante del manejo de una plantacion.

While the demand for big-leaf mahogany and Spanish cedar timber does not decline, the supply of these regionally important fine hardwood species has dwindled significantly. Plantations can help meet this demand, if the shoot borer is controlled through silvicultural techniques as well as selected seed sources. A participatory domestication program was implemented to improve germplasm quality and access in the Yucatan peninsula. Medium and short term strategies included the establishment of provenance/progeny trials, the conservation of a natural seed stand, and selection of urban seed trees. Five and two year data show significant differences between seed sources for growth and pest attack characteristics, indicating potential for genetic and economic gain. The process used for selection of seed trees in a 400 ha forest was simple and inexpensive, but also demonstrated that many trees may not be suitable for seed sources. The use and comparison of seed from urban trees is a useful educational tool in the domestication process. All of these efforts show the importance of a wide genetic base in reforestation programs. As with other timber species, specific steps can be taken to improve germplasm quality, and thus economic livelihood of people who plant trees.

One concern in the ongoing debate over the conservation status of mahogany has been the possibility that selective logging, as a form of negative phenotypic selection, might have led to deterioration in the genetic quality of populations. The incidence and degree of such ‘dysgenic’ effects is discussed, based on a consideration of mahogany logging practices, their expected genetic effects and empirical data on phenotypic selection in forest trees. Loggers have tended to avoid diseased, small, very large and poorly-formed individuals, and consequently logging has tended to increase the proportion of poor quality phenotypes relative to that before logging; in at least some conditions, selection differentials have been strongly negative. However, the upper limit for heritability of logger-selected traits in naturally-regenerated mahogany is probably no more than c . 0.1. Consequently, and assuming relatively extreme but realistic negative selection differentials of 50%, the maximum negative dysgenic response to a single logging-mediated phenotypic selection event is expected to be relatively small, i.e. ≤5%. This expectation is consistent with the empirical information from mahogany and other taxa. The implications of any dysgenic effects depend very much on the use and the future of dysgenically-selected populations. In managed populations, dysgenic effects could be reversed through positive selection. In the case of exploited but currently unmanaged natural populations, dysgenic selection is primarily of importance insofar as it affects fitness. As a threat to mahogany conservation and long-term sustainable production, it is probably insignificant in comparison with other genetic and non-genetic factors.