There are four different treatments on which conditional distribution.
The marginal p.d.f. of treatment Y is given in the bottom row of the referring table in the text.
The conditional p.d.f. of the response given each treatment is the ratio of the two rows above that to the bottom row.
\({g_1}\left( {x\left| 1 \right.} \right) = \left\{ {\begin{array}{*{20}{c}}{\frac{{0.120}}{{0.267}} = 0.449438}&{if\,x = 0}\\{\frac{{0.147}}{{0.267}} = 0.550562}&{if\,x = 1}\end{array}} \right.\)
\({g_1}\left( {x\left| 2 \right.} \right) = \left\{ {\begin{array}{*{20}{c}}{\frac{{0.087}}{{0.253}} = 0.343874}&{if\,x = 0}\\{\frac{{0.166}}{{0.253}} = 0.656126}&{if\,x = 1}\end{array}} \right.\)
\({g_1}\left( {x\left| 3 \right.} \right) = \left\{ {\begin{array}{*{20}{c}}{\frac{{0.146}}{{0.253}} = 0.577075}&{if\,x = 0}\\{\frac{{0.107}}{{0.253}} = 0.422925}&{if\,x = 1}\end{array}} \right.\)
\({g_1}\left( {x\left| 4 \right.} \right) = \left\{ {\begin{array}{*{20}{c}}{\frac{{0.160}}{{0.227}} = 0.704846}&{if\,x = 0}\\{\frac{{0.067}}{{0.227}} = 0.295154}&{if\,x = 1}\end{array}} \right.\)
Therefore, the fourth one looks quite different from the others, especially from the second.
