Abstract: Lotus seeds have been manually shelled mostly, because conventional shelling equipment is complexly structured and high pieced, and generally it has a high operating cost and a low reliability, as well as is difficult to be popularized. In this study, a new lotus seed sheller with simple structure and easy operation was developed to improve the reliability, to popularize the shelling mechanization of fresh lotus seed, and to reduce the burden of the farmers. When the newly designed sheller is running, the lotus seeds in the hopper fall into the evenly distributed grooves of the discharge wheel and are discharged singly into the feeding chute. As rolling down from the chute under the action of gravity, the lotus seeds are then normally positioned and fed into the cutting channel at a position where the long axis of the seed is perpendicular to the cutting edge. An annulus cut is made at the equator of the seed with the use of a reciprocating cutting blade. Thereafter, the shell and the kernel of each seed are separated by rubbing the seed surface with the pressing plate and discharged. The cutting device is the primary working device in lotus seed shelling. A comparison study on the effects of shelling methods was conducted. The 4 shelling methods used were: 1) The cutting blade was fixed with the pressing plate; 2) The cutting blade was fixed with the pressing plate but vibrated; 3) The cutting blade reciprocated and the pressing plate had a smooth surface; 4) The cutting blade reciprocated and the pressing plate surface was non-slip. The testing results indicated that the shelling method using a reciprocating cutting blade and a non-slip pressing plate was superior to the others in both shelling rate and shelling quality, which resulted in a shelling rate of 92%, a whole kernel rate of 100%, and a percentage of kernels with slight injury of less than 10%. The shelling method with reciprocating cutting blade and non-slip pressing plate could meet the shelling requirements of fresh lotus seeds with an equatorial diameter ranging from 14 to 23 mm. In order to get a better understanding of the factors closely related to the lotus seed processing quality, the mechanical properties of fresh lotus seed were measured using a universal testing machine (Model 5543, INSTRON, United States of America) in College of Biosystems Engineering and Food Science, Zhejiang University, China. The test head moved at a speed of 20 mm/min and stopped when the pressing force dropped by 15%. The samples of 100 seeds from the variety TAIKONG36 were randomly selected and numbered. The cutting force was tested firstly using the prototype blade. Then the same samples were shelled and subjected to the kernel firmness measurement using a 2.2 mm test head. The kernel firmness of each seed was measured right after shelling to minimize the effect of seed moisture loss. The results showed that the average cutting force of the shell was 19.63 N with a maximum value of 25.80 N. The average kernel firmness was 19.02 N with a maximum value of 23.27 N. The average kernel firmness per unit area was 5.00 N/mm2 with a maximum value of 6.12 N/mm2. Accordingly, the structure parameters and working parameters of the sheller were determined.