The Ordos Plateau of North China has experienced serious desertification. Hedysarum leave, Caragana korshinskii, Artemisia sphaerocephala and A. ordosica are four dominant psammophyte species that inhabit the Mu-Us Sandland, Ordos Plateau that are used in aerial seeding, a primary method for vegetation restoration in deserts and sand lands, in the Ordos Plateau, Seedling emergence is closely related to sand burial depth and water content of the soil but survival is low. In this paper, seed germination and seedling emergence characteristics of the four species in response to different levels of water availability and burial depth were studied and compared to determine the optimal time for aerial seeding to enhance seedling emergence and improve vegetation restoration efforts. The experiment was conducted in a non-heated greenhouse at the Ordos Sandland and Grassland Ecological Station of the Chinese Academy of Sciences in the summer of 2003. All seeds used in the experiment were collected from their natural habitat near the ecological station. The seed mass (mean±SE) varied among species: H. leave, (11.70±0.9) mg; C. korshinskii, (34.15±2.2) mg; A. sphaerocephala, (0.67±0.12) mg; A. ordosica, (0.22±0.09) mg. Seed germination, tested under the same environmental conditions, also varied among species: H. leave, (65.6±4.8)%; C. korshinskii, (52±4.7)%; A. sphaerocephala, (94.4±1.6)%; A. ordosica, 95.2%±0.8%. Each species was grown at seven sand burial depths (0, 0.5, 1, 1.5, 2, 3, 5 cm) under three water supply regimes (123, 185, 246 ml). The different water treatments corresponded to mean monthly precipitation levels of 50, 75, and 100 mm during growing season (from June to September). The average monthly precipitation in June in this area was about 50 mm. There were 21 treatments for each species with 5 replicates of each treatment. Each replicate consisted of 25 seeds planted in a plastic pot 5.6 cm in diameter and 11 cm in height with a drainage hole at the bottom of pot. Seedling emergence was checked and recorded everyday and water was added every three days. The experiment was terminated after 30 days when the seedling emergence became almost steady. The number of seedlings that did not emerge was quantified. As the main purpose of this study was on seedling emergence, dormant and decomposed seeds were counted as non-germinated. For each species, sand burial depth, water supply regime, and their interactions all had significant effects on the percentage and the rate of seedling emergence. All seeds lying on the soil surface did not germinate. The optimal burial depth for H. leave, C. korshinskii, A. sphaerocephala and A. ordosica was 0.5, 1, 0.5, and 0.5 cm, respectively. As the sand burial depth increased, the percentage and rate of seedling emergence decreased and the emergence time was delayed, but the percentage of seedlings that did not emerge was not affected. Seeds of A. sphaerocephala and A. ordosica could not emerge from burial depths greater than 1.5 cm, but seeds of H. leave and C. korshinskii were able to emerge from depths of 2-3 cm. This difference may be because of the greater mean seed mass of H. leave and C. korshinskii. The tolerance to sand burial depth was in the following order: H. leave> C. korshinskii > A. ordosica> A. sphaerocephala. The optimal water supply for all four species was 123 ml, which was close to the mean monthly precipitation of June (50 mm), the time when most seedling emergence occurs in the Ordos Plateau. As the amount of water increased, the percentage and rate of seedling emergence decreased and the emergence time delayed, but the percentage of seedlings that did not emerge was not affected. The maximum percentage (mean±SE) of seedling emergence of these four species was H. leave, 56.80%±1.96%, C. korshinskii, 39.20%±5.12%; A. sphaerocephala, 62.4%±4.12%; and A. ordosica, 77.6%±9.85%. Too much water and too deep of burial reduced the percentage of seedlings that emerged. This reflected the adaptation of the plants to the local climate and environment. In the Ordos Plateau, aerial seeding is often conducted during early June. Based on our results, weather conditions are not suitable for seedling establishment during this time because wind speeds are low, and the seeds are not buried by the sand. Our results indicate that this may reduce the percentage of seedlings that emerge as all four species require sand burial. During middle and late May, the winds are stronger and the seeds have a greater chance of being buried at their optimal depth. Even though there is no precipitation or the seeds might be buried too deep, they can survive in the soil and emerge later under more favorable conditions. Therefore, it is suggested that the time of aerial seeding be advanced from early June to mid to late May in order to enhance the emergence of the seedlings.