My code are follows:



\Tree [.CauseP [.DP \edge[roof]; {`the zog'} ] [ [.{\scshape
cause} ] [ [.{\scshape past} ] [.vP \edge[roof]; {`dissolved the
house'} ]]] ]

Throughout this monograph, I will
focus on the reusable linguistic ingredients of the system and how
they are efficiently deployed in the recursive combinatorics in
building propositions. Reusability (and by extension polysemy)
will be a seen as a design feature of the system rather than a
bug. As a methodological principle, then, I will be guided by the
fact of polysemous items and seek analytic unities to underlie

\ex. \textit{Achievements and accomplishments with result states}\\
%\begin{tikzpicture}[baseline=(current bounding box.north)]
\begin{adjustbox}{max width=0.91\textwidth}
\Tree [.InitP [.{\textsc{initiator}} ] [ [.Init ] [.ProcP
[.{\textsc{undergoer}} ] [ [.Proc ] [.ResP
[.{\textsc{resultee}} ] [ [.Res ] [.XP \edge[roof];
{\sc ground/final state} ] ] ] ] ] ] ]


and the output came as:

enter image description here

Is this possible to make the diagonal lines to be straight as marked in the below picture?

enter image description here

Diagonal line should be straight in all Trees. Please suggest...

  • Similar question for forest: tex.stackexchange.com/questions/391820/… – Torbjørn T. Dec 13 '17 at 13:20
  • @TorbjørnT. That case is rather more complicated. I think nice empty nodes and an adjustment to the edge angles is all that's needed for a Forest solution in this case. There is probably a better link, but I've posted an illustration here rather than trying to find it :-). – cfr Dec 13 '17 at 23:43
  • \sc is deprecated by more than a couple of decades and ought not be used with LaTeX 2e. – cfr Dec 13 '17 at 23:44
  • 1
    @cfr On the more general forest solution, I think A nice empty node with nice nodes in forest? is appropriate. But since this question is about tikz-qtree specifically, I think it's a duplicate of "Kinks" in long tikz-qtree branches. – Alan Munn Dec 14 '17 at 1:12

Forest can do this - sort of - if that's an option. However, Forest's trees tend to be rather more ungainly by default, at least in a range of pertinent cases.

Here's an example using nice empty nodes:

<code>nice empty nodes</code>

  nice empty nodes,
  for tree={
    calign angle=50,
    [DP[`the zog', roof]]
    [[cause, scshape]
      [[past, scshape][vP[`dissolved the house', roof]]
  • Many thanks to all, its working fine, but is this possible to make the requirement without using the forest.sty, please suggest.... – MadyYuvi Dec 14 '17 at 6:47
  • @MadyYuvi Yes. Just use the solution Alan Munn linked you to for the package you're using. Alan's answer there provides solutions for both packages. – cfr Dec 16 '17 at 4:48

Not the answer you're looking for? Browse other questions tagged or ask your own question.