Approximately 26% of the coffee grown in Colombia is located in areas presenting water deficit, with some of these areas also presenting high solar brightness. This combination reduces coffee production, thus affecting the income of 31% of the country's coffee-growing families. To identify accessions of the Colombian Coffee Collection (CCC) that are tolerant to water deficit, 65 genotypes were evaluated in screenhouse conditions at the National Coffee Research Center (Cenicafé), located in Manizales, Caldas, Colombia. Seedlings of each genotype were transplanted to polyethylene bags, each filled with 10 kg sandy loam Andisol soil. Two moisture treatments were applied as follows: (1) soil at field capacity (60% moisture) and ( 2) water deficit conditions, with soil at 50% field capacity (30% moisture). After five months, total dry biomass was determined, considered as the sum of the dry biomass of leaves, stems, and roots. The Student's t test for independent samples was used to analyze resulting values at a level of significance of 5%. Reducing irrigation under water deficit conditions usually delays accession growth, which is reflected in decreased biomass. However, the total dry biomass of nine Ethiopian introductions of Coffea arabica (CCC238, CCC254, CCC284, CCC372, CCC474, CCC536, CCC537, CCC555, CCC1147), six diploid accessions (CCC1030, EA.20, EA.209, EA.227, EA.229, EA.287), and three interspecific hybrids of Caturra x Coffea canephora (25, 640, 702) in water deficit conditions did not differ statistically from the total dry biomass obtained in treatments with irrigation at field capacity. Because these introductions present adaptation mechanisms to water deficit, they retain their leaves without reducing their leaf area or total dry biomass and should accordingly be considered as candidates for evaluation in dry regions to determine their tolerance to water deficit based on effects on production or biomass.