The cassava plant (Manihot utilissima Pohl) is native in Brazil and was already used by the indians as a main source of food before the arrival of the Europeans. Wild species of Manihot are found in several parts of Brazil and in other South American countries. Cassava roots are one of the chief sources of carbohydrates for a large part of the Brazilian population. It is also widely used to feed animals. In parts of Brazil they are used as raw material for the starch industry. In 1946 there were about 900 000 hectares of cassava planted in Brazil. In the State of São Paulo little attention had been given to the agricultural problems of the cassava crop prior to 1935. At this time an extensive series of field trials was started by the Instituto Agronômico, Campinas, to obtain information on the various problems of this crop. Since then a large amount of data has been obtained, and the present paper is a summary of the work that has been carried out. Over 600 varieties were collected from several parts nf Brazil to be used as basic material for selection or synthesis of high yielding and disease resistant types. Extensive field tests have shown that the highest yielding and most resistant varieties were: "Branca de Santa Catarina", "Preta", "Cafelha", "Brava de Itu", and "Itu". They are more suitable for industrial purposes than the common variety known as "Vassourinha". The variety "Guaxupé" was found to be the best for human consumption and as a forage. In experiments on methods of planting, the use of cuttings 50 cm long stuck in the soil in a vertical or slanting position, leaving about 35 cm above ground, gave better results than cuttings 15 cm long, placed in a horizontal position in furrows as it is usually done by the farmers. Fertilizer experiments showed that, as a rule, phosphates gave the highest increase in yield. Nitrogen or potassium, either alone or together, had no effect on yield. When one or both of these elements were added to phosphates the yields were better than with phosphates alone. The period from May to August is the regular cassava harvesting time in the State of São Paulo. Experimental plantings made during these months gave higher yields than those made in October, the usual time of planting. Furthermore, when the planting is made soon after harvest the losses of cutting that occur during a long storage are avoided. In field test cuttings 20 to 25 cm long, planted in furrows, secured better stands and higher root production than were obtained with shorter cuttings. Data from field trials indicated that a spacing ranging between 0,80 x 0,40 cm and 1,00 x 0,60, according to soil fertility, is more advantageous than the spacing of 1,20 x 0,60 cm as it is usually adopted. Test on depth of planting were made with 15 cm long cuttings planted in furrows 5, 10 and 15 cm deep, and subsequently covered. The plants grown from cuttings planted 15 cm deep produced less and were more difficult to be dug out. Planting at 5 cm depth is also inadvisable because the plants may be easily uprooted by erosion or strong winds. Planting at 10 cm depth is to be recommended. Studies on cassava bacterial wilt caused by Xanthomonas manihoti (Arthaud-Ber-thet) Burk, revealed that several common varieties and clones derived from seedlings show more resistance than the usually cultivated types. Cuttings of the resistant types were released to the growers and are now being widely used. A virus disease of the witches broom type, present in some localities, caused severe losses in cassava plantings made with the variety "Vassourinha". Tests carried out in infested areas showed that the variety "Brava da Ponte" is highly resistant to the disease, and the varieties "Preta" and "Holandi do Itaguá" show a fair degree of resistance.