Global change is potentially expected to change the quality and quantity of aboveground litter input, which could affect belowground processes of terrestrial ecosystems. Root proliferation into the litter layer in subtropical forests substantially contributes to net primary production; however, little is known concerning how increased leaf litter input affects the growth of fine roots. A litter manipulation experiment was conducted between August 2013 and August 2014 in nine stands of three different forest types in subtropical China: coniferous forest (CF), coniferous and broad-leaved mixed forest (MF), and evergreen broad-leaved forest (BF). Three treatments, including litter removal (LR), litter addition (LA), and a control (CK) were conducted within 5 m × 5 m plots in a randomized block design in each stand. The biomass of fine roots (≤2 mm in diameter) in the litter layer and the 0-15 cm soil layer were measured using soil cores; root morphology was also monitored. The results showed that litter removal significantly decreased fine root biomass in CF and MF plots by 40.3% and 37.5%, respectively. Litter addition caused a slight decline in fine root biomass in CF and MF plots, but fine root biomass markedly increased (19.4%) in BF plots. The vertical distribution of fine roots differed according to litter treatments. Because CF to BF correspond to a successional gradient and increasing tree species diversity in subtropical forests, forest floor litter mass and thickness in the litter horizon increased, as did the effects of litter quantity on the vertical distribution of fine roots. Compared to the controls, litter addition plots in the relatively fertile soil of BF, resulted in a significantly higher percentage of fine roots in the litter layer, but a significantly lower percentage at 7.5-15 cm depth in the mineral soil (increase of 10.6% and decrease of 10.4%, respectively). This suggested that the results were a response to a more readily available nutrient source rather than an adaptation to nutrient shortage. Root biomass in the litter layer in the CK and LA plots was strongly related to litter layer depth (R² = 0.742, P < 0.01, linear regression) and total litter biomass (R² = 0.521, P < 0.01, linear regression). Moreover, root length density and specific length of fine roots differed between substrate layers: specific root length was greater in the litter layer than in the mineral soil layer, and root length density was the greatest in the 0-7.5 cm soil layer. However, fine root morphology did not change significantly with addition or removal of litter. Therefore, we suggest that fine roots have a nutrient acquisition strategy that allows uptake of more nutrients from the litter layer or near the surface in shallow soils because of a larger belowground allocation of biomass and a more active metabolism, not because of phenotypic plasticity in fine root morphology.