The data set I am working with has spaces. Structure: SUBURB TOWN NUM1 NUM2 Since Suburbs and Towns often have Spaces in i have chosen to use the typical CSV structure. VALLEY VIEW,JONES TOWN,2032,2033 Using the read command I can easily save it into variables using the dataset but the issue is that read -a defaults to a space character as the delimiter, so the above becomes... ${variable[0]} is VALLEY ${variable[1]} is VIEW,JONES ${variable[2]} is TOWN,2032,2033 ${variable[3]} is null I need to use the comma character as the delimiter so I can get the following... ${variable[0]} is VALLEY VIEW ${variable[1]} is JONES TOWN ${variable[2]} is 2032 ${variable[3]} is 2033 I tried using ... read -d "," -a variable <<< $dataset No difference. Any one got some idea?