The development and progress of antidiabetic drugs (e.g., insulin preparations and hypoglycemic drugs) are retrospectively investigated in Japan. Their influences on the treatment of diabetes mellitus (DM) and its epidemiological aspects are also discussed. 1) Insulin preparations: Insulin was introduced for DM therapy in 1925, two or three years after its discovery in Canada. The preparations were raw extracts of bovine or porcine pancreas. These did not prevail widely in Japan because of the low incidence of DM before World Wan II. After the war, a shortage of mammalian materials compelled the use of fish pancreatic tissues such as bonito and/or tuna for insulin production. Insulin infection, so-called regular insulin, was first promoted in the 6th "Pharmacopoeia Japonica" (JP6) in 1951 and has been maintained to the present edition (JP14, 2001). Although depot-type insulin preparations were developed in the USA and Europe during the war, the introduction of those preparations to Japan was delayed until 1951, when Protamine zinc insulin appeared. Globin zinc insulin and Isophane insulin were introduced for clinical use in 1952 and 1955, respectively. These were also adopted for JP7 (1961). Biphasic-type insulin, which has a rapid onset and long duration of activity, appeared in 1965. Purified preparations from bovine or porcine sources have been available since 1980, which might be a strong reason for the decrease in insulin allergy. Insulin from animal origin has been supplied for almost 60 years since its discovery. Amino acid sequences of insulins from various species of animals were determined by the pioneering studies of Sanger and his associates. Human insulin, which differs from porcine insulin by only one amino acid, was produced by Novo researchers in 1982 using a semi-synthetic method. Then the Lilly group soon succeeded in obtaining human insulin by recombinant DNA technology in the same year. Both products were introduced to Japan in 1985, and the recombinant products prevailed throughout the 1990s. Human insulin analogues (i.e., Insulin lispro and Insulin aspart) appeared in 2001. These are applied for after-meal glycosmia owing to their ultrarapid onset of activity. Self-injection by DM patients was legalized in 1981. To make the infection technique sure and easy, cartridge (pen-type) and disposable kit-type needles were devised in the 1990s. 2) Oral hypoglycemic drugs: Instead of the exclusive parenteral usage of insulins, there was also demand for oral dosage forms. The first of the sulfonyrlurea (SU) group, BZ-55, was used for DM clinically in 1955 in Germany. But it was soon withdrawn because of its antibacterial action. This led to the development of various SU groups. Tolbutamide (1956), chlorpropamide (1959), acetohexamide (1964) and tolazamide (1961) were introduced to Japan as first-generation SUs. Then glyclopyramide (Kyorin, 1965), glybenclamide (1971), gliclazide (1984) and glimepiride (1999) appeared as the second-generation SUs. These were used orally for Type 2 diabetes. Biguanide (BG) group, phenformin HC1 (1959), metformin HC1 (1961) and buformin HC1 (1961) had also been in use by oral treatment of Type 2 diabetes. SU appears to act by increasing the sensitivity of b-cells, which secrete insulin. BG probably exerts by increasing glucose transport across the membranes of target organs. 3) New types of antidiabetic drugs: a-Glucosidase inhibitors (i.e., acarbose: Bayer, 1993; and voglibose: Takeda, 1994) act on hyperglycemia after meals by decreasing glucose absorption. Thiazolidinedione compounds, such as troglitazone (Sankyo, 1995) and pioglitazone HC1 (Takeda, 1994) act by increasing the insulin sensitivity of the target tissues. These are useful for Type 2 DM patients when SUs are ineffective. Nevertheless, troglitazone was discontinued in 2000 due to severe liver damage. Nateglinide (Ajinomoto Co., 1999), which is a D-phenylalanine derivative acting similar to SUs, is useful orally for after-meal hyperglycemia of Type 2 diabetes. Epalrestat (Ono Yakuhin Co., 1992) is effective for diabetic neuropathy by reducing the formation of sorbitol. These anti-DM drugs were recently studied and developed in Japan. 4) The Japan Diabetes Society proposed a guideline on diagnostic criteria and treatment of diabetes mellitus (DM) in 1999 and revised it in 2002. DM is classified as insulin-dependent DM (Type l) and non-insulin dependent DM (Type 2). Type 1, juvenile onset DM, requires insulin therapy to prevent ketosis and to sustain life. Treatment of type 2, adult onset DM, is recommended as a step-by-step method, starting with dietary-exercise therapy, followed by oral hypoglycemic drugs and then insulin therapy. DM patients with complications should have a therapy devised to match their circumstances. 5) Epidemiological aspects: The mortality rate of DM compared to the time of drug appearance was traced from 1920 to 2000. The curve goes down slowly in the time frame of World War II, but rises from 1950 to 1970. The elevation could not be suppressed by the appearance of SUs, BGs or improved insulin preparations. The curve runs flat from 1980 to 1990, which might be related to the use of purified insulin or human insulin therapy. The mortality rate of DM indicates that death by hyperglycemic coma and other deaths resulting from complications are excluded. The survey of the principal cause of death by DM during the period of 1981-1990 indicates that the death rate due to hyperglycemic coma is only 1.7% of the total deaths caused by DM. The effect of drug therapy on all of the death resulting from DM is not detected. Hospital visitation and admission rates of the DM patients have been recorded since 1952 in Japan. This curve is rising continuously, and none of the antidiabetic drugs has been able to suppress it. These data show that the antidiabetic drugs relieve DM symptoms through their effective hypoglycemic actions, but that they cannot suppress the mortality rate of DM. It is possible that none of the drugs currently available can suppress the increasing tendency of DM patients.