Many herbicides that are applied at the soil before weed emergence inhibit plant growth soon after weed germination occurs. Plant growth has been known as an irreversible increase in size as a result of the processes of cell divison and cell enlargement. Herbicides can influence primary growth in which most new plant tissues emerges from meristmatic region by affecting either or both of these processes. Herbicides which have sites of action during interphase($G_1$, S, $G_2$) of cell cycle and cause a subsequent reduction in the observed frequency of mitotic figures can be classified as an inhibitor of mitotic entry. Those herbicides that affect the mitotic sequence(mitosis) by influencing the development of the spindle apparatus or by influencing new cell plate formation should be classified as causing disruption of the mitotic sequence. Sulfonylureas, imidazolinones, chloroacetamides and some others inhibit plant growth by inhibiting the entry of cell into mitosis. The carbamate herbicides asulam, carbetamide, chlorpropham and propham etc. reported to disrupt the mitotic sequence, especially affecting on spindle function, and the dinitroaniline herbicides trifluralin, nitralin, pendimethalin, dinitramine and oryzalin etc. reported to disrupt the mitotic sequence, particularly causing disappearence of microtubles from treated cells due to inhibition of polymerization process. An inhibition of cell enlargement can be made by membrane demage, metabolic changes within cells, or changes in processes necessary for cell yielding. Several herbicides such as diallate, triallate, alachlor, metolachlor and EPTC etc. reported to inhibit cell enlargement, while 2, 4-D has been known to disrupt cell enlargement. One potential danger inherent in the use of soil acting herbicides is that build-up of residues could occur from year to year. In practice, the sort of build-up that would be disastrous is unikely to occur for substances applied at the correct soil concentration. Crop injury caused by soil applied herbicides can be minimized by (1) following the guidance of safe use of herbicides, particularly correct dose at correct time in right crop, (2) by use of safeners which protect crops against injury without protecting any weed ; interactions between herbicides and safeners(antagonists) at target sites do occur probably from the following mechanisms (1) competition for binding site, (2) circumvention of the target site, and (3) compensation of target site, and another mechanism of safener action can be explained by enhancement of glutathione and glutathione related enzyme activity as shown in the protection of rice from pretilachlor injury by safener fenclorim, (3) development of herbicide resistant crops ; development of herbicide-resistant weed biotypes can be explained by either gene pool theory or selection theory which are two most accepted explanations, and on this basis it is likely to develop herbicide-resistant crops of commercial use. Carry-over problems do occur following repeated use of the same herbicide in an extended period of monocropping, and by errors in initial application which lead to accidental and irregular overdosing, and by climatic influence on rates of loss. These problems are usually related to the marked sensitivity of the particular crops to the specific herbicide residues, e.g. wheat/pronamide, barley/napropamid, sugarbeet/ chlorsulfuron, quinclorac/tomato. Relatively-short-residual product, succeeding culture of insensitive crop to specific herbicide, and greater reliance on postemergence herbicide treatments should be alternatives for farmer practices to prevent these problems.