As habitat loss increases, the survival of certain valuable species, such as the giant panda and the tiger, is becoming increasingly challenged. Nature reserves are attracting increased attention as tools for protecting valuable endangered species. When designing nature reserves, certain important issues, including whether several large or many small reserves are optimal-which represents the famous few larger or many small debate (FLOMS) -must be considered. In a fragmented landscape, we selected a region from the center of the landscape, so that the area of suitable habitats in the region is fixed and constant. As such, the selected region should be larger if the habitat loss is more severe. Subsequently, we explored whether several large or many small reserves should be distributed over these suitable habitats. The results suggest that:(1) for random habitat loss, when the proportion of lost habitats is 0.2, the optimal reserve number is more than 170. As the proportion of lost habitats increases, the optimal reserve number also decreases sharply, and when it reaches 0.9, the optimal reserve number is less than 20. Because the area of every reserve is in inverse proportion to the number of reserves, when random habitat loss occurs, increases in the proportion of lost habitats tend to favor the implementation of several large reserves. (2) If the proportion of lost habitats is fixed and the degree of clustering of lost habitats is low, then the optimal reserve number is small. As the degree of clustering increases, the optimal reserve number will also increase sharply, and when it is high, the optimal reserve number will also be high. As a result, although the proportion of lost habitats is constant, the spatial distribution of lost habitats also greatly affects the FLOMS problem. In addition, increases in the degree of clustering of lost habitats tend to favor the implementation of many small reserves. (3) When the diffusion rate was 0, the optimal reserve number is also low. As the diffusion rate increases, the optimal reserve number also increases. If the degree of clustering of lost habitats is higher, then a greater increase in the optimal reserve number occurs under an increasing rate of diffusion. When the diffusion mortality rate is 0.9, the optimal reserve number is low. As the diffusion mortality rate decreases, the optimal reserve number also increases. If the degree of clustering of lost habitats is higher, then a greater increase in the optimal reserve number will occur with a decreasing diffusion mortality rate. As a result, increases in the diffusion rate or decreases in the diffusion mortality will tend to favor the implementation of many small reserves. When the degree of clustering of lost habitats is higher, a greater increase in the optimal reserve number will occur with increasing or decreasing diffusion rate. These findings can provide a theoretical basis for the FLOMS debate and endangered species conservation, and reinforce the importance of habitat loss, providing insights for natural environmental protection.