Abstract
Tubers of cassava are used as staple food, feed, and raw material in industries. Field and laboratory
experiments were conducted during 2009–‘2012 under subtropical condition (24°75' N 90°50' E) to
characterise cassava accessions based on important morphological characters, nutritional
composition, and hydrogen cyanide (HCN) content and its elimination, growth and yield. Results
across field and laboratory experiments showed significant genetic variation in canopy characters,
proximate composition (crude protein, crude fat, crude fibre and ash), starch and HCN contents, and
tuber yield. Eleven cassava accessions were screened based on morphological descriptors,
proximate compositions of leaf and tuber (root), biomass of root and shoot. Stem scar, petiole
colour, leaf-lobe character and external colour of tuber contributed to most of the variations for
identification of accessions. Based on tuber and starch yields, and protein and HCN contents two
accessions (Coc-A1 and Kha-A2) were selected for detail agronomic studies with the former had
higher yield and protein but lower HCN content. A number of cultural practices were employed:
number of shoots allowed (one vs two) from cutting, two positions of mainstem cutting (proximal
vs distal) placement, three depths of planting (6, 12 and 18 cm), three methods of planting
(horizontal, lateral and vertical), and six dates of planting (Jan, Feb, Mar, Apr, Nov and Dec) were
used in the above two selected accessions. Results revealed that single shoot allowed from proximal
region of the mainstem cutting with 12 cm depth of planting and horizontal method of cutting
placement in the soil using Nov or Feb date of planting produced higher starch, tuber and biomass
yields with the magnitude being greater in the Coc-A1 than in the Kha-A2. In cassava, HCN is
naturally produced from plant parts during postharvest processing like chopping, boiling, and
drying. Elimination of HCN by processing techniques viz., effect of boiling (100°C) and drying
(50°C) on HCN removal was investigated. Irrespective of accessions, it appeared that fresh plant
parts contained a high quantity of HCN (133.30 –531.50 mg kg™ fresh tissue) which was deadly
harmful (lethal dose for human > 50 mg HCN 50 kg¹ body wt); and boiling and drying removed
99% HCN i.e. to a safe level. Starch and tuber yields were also investigated 2, 4, 6, 8, 10, 12 months
after planting (MAP) in the two accessions, Coc-A1 and Kha-A2.It was further observed that higher
starch (Ave 28.36% FW) and tuber yields (Ave 45.77 t ha¹ FW) were observed during 8 to 10
MAPs with the magnitude being again greater in the former accession. Results concluded that
cassava accessions could be identified by morphological descriptors, growth, nutritional
composition and yield. HCN produced naturally in cassava plants during postharvest operation
could be removed to a safe level by processing. Optimal protocols for realisation of maximum
biomass and yield were also obtained.