The emissions of diesel particulate matter (DPM) from diesel engines are causing increasing health concerns due to their suspected carcinogenicity. DPM consists mostly of carbonaceous materials, which is often classified as elemental carbon (EC) and organic carbon (OC). EC and OC affect the environment in multiple ways due to their optical, physical, chemical, and toxicological properties. Sulfur species are also one of the important components in non-road diesel emissions. Especially, sulfate (SO4(2-)) is supposed to play an important role in particulate formation and organic compounds nucleation. This research investigated the distributions of carbon and sulfur speciation under various source conditions, and the study was performed on a non-road diesel generator. For the study of carbon speciation, samples were colleted using a EPA Method 5 sampling train, and OC/EC in DPM were measured by the thermal-optical method NIOSH 5040. Tests were performed at various engine loads, and the influence of diesel sulfur content and collection temperature on the OC/EC distributions were discussed. Results showed that DPM concentrations and the relative contributions of OC, EC, and other unaccounted mass vary greatly with engine load, fuel sulfur content, and sample collection temperature. The EC concentrations in DPM exhaust increase with increasing load, while the OC concentrations do not show great variation with load. It is also found that in the high sulfur diesel emissions both the OC and non-carbonaceous materials contribute more to DPM than in the low sulfur diesel emissions. As expected, the collection temperature showed no influence on EC concentrations while it has a great influence on the OC and non-carbonaceous materials. For the study of sulfur speciation, EPA Method 8 was utilized. DPM samples were collected at the same time as measuring sulfur dioxide (SO2) and sulfuric acid (H2SO4) in emissions. Particulate sulfate (SO4(2-)) and Total particulate sulfur in DPM were determined by Ion Chromatography (IC) and X-Ray Fluorescence (XRF) spectroscopy analysis respectively. Results showed that, SO2 concentration is clearly related to diesel sulfur content as well as engine load conditions in non-road diesel emissions. The S-->SO2 conversion rates slightly decrease with increasing load. And they are obviously related with diesel types. For the same type of diesel, the conversion rates increase as diesel sulfur content increase. However, the total SO4(2-) concentration is not sensitive to diesel sulfur content. It is also found that the sulfur recovery is sensitive to diesel fuel types and may be related to the forms of sulfur compounds in diesel fuel. Besides the study of carbon and sulfur speciation in DPM, an investigation of sampling artifacts was performed on a high volume sampling system. Quartz pair tandem filters were used in the tests, and DPM influenced samples were collected at two different filter face velocities and a series of collection time. It is found that, under the specified sampling conditions in the study, OC on the backup filter was mainly from adsorption of vapor but not from volatilization of particles on front filter. The quartz-quartz pair tandem filter method can be used to correct the adsorption artifacts by subtracting backup filter measurements from the front filter measurements. However the accuracy of the method improves with increased collection time. In order to effectively correct the adsorption artifacts, collection time should be long enough to approach gas/filter adsorption equilibrium. The study also estimated the capability of the applied quartz filter to adsorb OC vapor and found that it was not obviously influenced by sampling face velocity. The information will be helpful to choose valid sampling duration time for field sampling in order to effectively correct sampling artifacts caused by adsorption.