I want to include a tikzpicture (mindmap) in beamer and I want the children to appear one after the other.

So I use \visible but it don't works -- also not with \pause command. I get errors regarding forgotten semicolons. When I insert semicolons, not all of the mindmap is created.

Is there an easy way to realize that?

\begin{tikzpicture}[mindmap, concept color=gray!50, font=\sf, text=white]

  \tikzstyle{level 1 concept}+=[font=\sf, sibling angle=90,level distance = 30mm]

  \node[concept,scale=0.7] {Gedächtnis}
    [clockwise from=135]}
        child[concept color=orange]{ node[concept,scale=0.7]{Musik} } }
        child[concept color=orange]{ node[concept,scale=0.7]{Kunst} } }
        child[concept color=orange]{ node[concept,scale=0.7]{Mathematik} } }
        child[concept color=orange]{ node[concept,scale=0.7]{Seltenere} } };


  • 1
    I'm not sure of there's yet a "proper" way to do this, but I've had good results with Matthew Leingang's very general solution in a previous question. This allows you to arbitrarily change properties per slide, color being the obvious target here. There may be some more hints in this previous thread as well.
    – kgr
    May 14, 2012 at 14:52
  • I think you should use \uncover instead of \visible.
    – uzsolt
    May 14, 2012 at 15:57
  • @kgr: Sorry, but I converted your post to a comment, because links to answers do not qualify as new answers. You could however add a full solution based on these answers. Simply edit and undelete your post. May 14, 2012 at 19:34

2 Answers 2


This answer has led to the aobs-tikz package by Claudio Fiandrino, which provides extended versions of the concepts below.

Here is a less hacky alternative to the answer given by Claudio:

In conjunction with TikZ, I have almost stopped completely to use beamer's overlay commands (\visible<>, \only<>, and so on). Instead I always draw all elements, but hide them on the slides they should not appear. To specify the visibility, I use a visible on=<...> TikZ style as follows:


  % Keys to support piece-wise uncovering of elements in TikZ pictures:
  % \node[visible on=<2->](foo){Foo}
  % \node[visible on=<{2,4}>](bar){Bar}   % put braces around comma expressions
  % Internally works by setting opacity=0 when invisible, which has the 
  % adavantage (compared to \node<2->(foo){Foo} that the node is always there, hence
  % always consumes space plus that coordinate (foo) is always available.
  % The actual command that implements the invisibility can be overriden
  % by altering the style invisible. For instance \tikzsset{invisible/.style={opacity=0.2}}
  % would dim the "invisible" parts. Alternatively, the color might be set to white, if the
  % output driver does not support transparencies (e.g., PS) 
    visible on/.style={alt={#1{}{invisible}}},
    alt/.code args={<#1>#2#3}{%
      \alt<#1>{\pgfkeysalso{#2}}{\pgfkeysalso{#3}} % \pgfkeysalso doesn't change the path

\begin{tikzpicture}[mindmap, concept color=gray!50, font=\sf, text=white]

  \tikzstyle{level 1 concept}+=[font=\sf, sibling angle=90,level distance = 30mm]

  \node[concept,scale=0.7] {Gedächtnis}
    [clockwise from=135]
        child[concept color=orange, visible on=<2->]{ node[concept,scale=0.7]{Musik} } 
        child[concept color=orange, visible on=<3->]{ node[concept,scale=0.7]{Kunst} } 
        child[concept color=orange, visible on=<4->]{ node[concept,scale=0.7]{Mathematik} } 
        child[concept color=orange, visible on=<5->]{ node[concept,scale=0.7]{Seltenere} };




visible on=< ovspec > is implemented by applying the style invisible on all slides that are not contained in ovspec. The default implementation of invisible just sets opacity=0; however, this can be overwritten, so other kinds of "invsibility" can easily be installed (dimming, gray filling, ...).

Note: If overspec itself contains a comma, either it or the complete argument has to be put inside curly braces (like in, visible on=<{1,3-4,8}> or visible on={<1,3-4,8>}) in order to not confuse the pgfkeys parser.

This approach (besides being, IMHO, much better readable) also has another advantage: All named elements (especially nodes) are always there, so you can use them for coordinate calculations even on slides they are not visible. Moreover, the tikzpicture always has the complete size, preventing the frequently asked issue of "jumping images".


  • 1
    @Dave: tikz and transparent probably clash in the backend driver. There have been some reported issues of using transparent and the opacity feature of tikz in conjunction. However, TikZ should provide you all the transparency features you ever need.
    – Daniel
    Jan 5, 2013 at 21:45
  • @Daniel: Ah, it was that simple?? I introduced some new styles visible2 on=... that would then convert spaces into commas. -_- Thanks for the solution!
    – gablin
    May 30, 2013 at 11:28
  • @Daniel: Actually, visible on={<1,3>} didn't work, but visible on=<{1,3}> did. =)
    – gablin
    May 30, 2013 at 11:38
  • @gablin: Right, it has to be that way, as obvious from the delimited argument specification. I have now edited this into the answer.
    – Daniel
    May 30, 2013 at 13:29
  • nice indeed. Is there a way to get it working with externalization (I'm using \usepgfplotslibrary{external} \tikzexternalize)? Oct 27, 2016 at 18:17

Here is a solution taken from this message and adapted to your need (just 1 level, clockwise option and position of concepts angle 90, but you will see that I commented the code where there is something taken from your needs).

Edit: seen now that I forgot the text=white; I will add a comment only in the place you should modify.



% excellent code taken from 
% http://old.nabble.com/Re%3A-successive-creation-of-tikz-mindmap-in-beamer-p22516097.html

\def\tikz@collect@child@@overlay#1[{\tikz@collect@childA[child overlay={#1},}


% Extra hackery to allow preactions on different layers.
      \path[#1];% do extra path
      \pgfsyssoftpath@setcurrentpath\tikz@actions@path% restore

\tikzset{preaction layer/.store in=\tikz@preaction@layer}


        child overlay/.code={%
                                circle connection bar switch color/.code={},
                                edge from parent/.style={draw=none},
                                every node/.style={
                                        concept, draw=none, fill=none,
                                        execute at begin node={\setbox0=\hbox\bgroup\hskip0pt\let\\=\relax},
                                        execute at end node=\egroup\phantom{\box0}


        use shadow/.style={%
                copy shadow={%
                        preaction layer=shadow, fill=gray!25, draw=none,
                        shadow xshift=0.5ex, shadow yshift=-0.5ex
        small mindmap/.style={
          level 1/.append style={level 1 concept},
          concept color=gray!50, % root colored in gray
          every concept/.style={align=center, font=\tiny\strut, text=black,% <= TEXT COLOR CHANGE
outer sep=-.25pt}, 
          text width=2cm,
          level 1 concept/.style={
          % font \sf as seen in your example
          text width=1.5cm,
          level distance=3cm,
          % some modifications to require needs
          sibling angle=90,
          clockwise from=135,
          every child/.style={concept color=orange},
          every node/.style={concept, execute at begin node=\hskip0pt, use shadow},
          every circle connection bar/.append style={append after
command={[use shadow]}}


\begin{tikzpicture}[remember picture, overlay, small mindmap]
\node <1->  {Gedächtnis}
                        child <2-> {node {Musik}}
                        child <3-> {node {Kunst}}
                        child <4->{node {Mathematik}}
                        child <5->{node {Seltenere}};


I show just three frames over the five generated:

enter image description here

enter image description here

enter image description here

  • Thanks for your solution. It's definitely not easy, but in principle it works. Only, when I try this, that is shifted over the left border. But it's not a problem because Daniel give a helpful solution too.
    – André
    May 16, 2012 at 7:39
  • Yes, you are right; the problem could be fixed by compiling twice. May 17, 2012 at 7:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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